WO2019078577A1

WO2019078577A1 - Methods and systems for assessing ovarian parameters from ultrasound images

Info

Publication number: WO2019078577A1
Application number: PCT/KR2018/012169
Authority: WO
Inventors: Ravi Teja NARRA; Srinivasan SIVANANDAN; Srinivas Rao Kudavelly; Nikhil Narayan SUBBARAO; Nitin Singhal
Original assignee: Samsung Medison Co., Ltd.
Priority date: 2017-10-16
Filing date: 2018-10-16
Publication date: 2019-04-25

Abstract

Embodiments herein disclose methods and systems for assessing an ovary of a subject for quantification of ovarian parameters by detection and segmentation of the ovary from an ultrasound scan of the ovary. Structures of the ovary are classified as either ovarian or non-ovarian based on location and type. The embodiments include determining and tracking the path through which needle is guided for oocyte aspiration. Ovarian reserve is estimated based on the quantified ovarian parameters. Ovarian cancer and PCOS can be diagnosed based on quantified ovarian parameters.

Description

METHODS AND SYSTEMS FOR ASSESSING OVARIAN PARAMETERS FROM ULTRASOUND IMAGES

Embodiments herein relate to obstetrics and gynecology, and more particularly to methods and systems for providing clinical aid in performing an assessment of female reproductive organs such as ovaries.

Ovary is an organ of the female reproductive system that is responsible for the synthesis of ovum. The ovum when fertilized develops into an embryo. The size of a normal ovary varies with age and its size can increase exponentially for approximately 20 years, after which it gradually reduces.

Quantification of the ovarian volume can aid in diagnosis and management of gynecological conditions such as infertility and cancer. Ovarian volume can be a useful indicator of the response to hyper-stimulation in assisted reproduction. There can be a direct correlation between the number of Non-Growing Follicles (NGF) and the ovarian volume. The dosage protocol for initiating hyper-stimulation can be determined by estimating the number of NGF based on ovarian volume. Based on the Rotterdam criteria, ovarian volume is one of the ultrasonographic indicators to classify the ovary as being poly-cystic or normal. Additionally, ovarian volume can also be considered as a useful marker for screening of ovarian cancer.

Ultrasound scans can be used for diagnosis of disorders in women's reproductive system, such as disorders that causes infertility (such as Poly-Cystic Ovary Syndrome (PCOS)), life threatening diseases (such as ovarian cancer), and so on. Such disorders can be detected manually. However, manual identification of these disorders can be time consuming, have low reproducibility, and high false negative rates. In an example, follicles of size 2-4mm may not be identified reliably in ultrasound scans.

FIG. 1 depicts anatomy of female reproductive system. As depicted in FIG. 1, the female reproductive system comprises of two ovaries, two fallopian tubes, the uterus, the serosa, the cervix, and the vagina. The ovaries are ductless reproductive glands wherein female reproductive cells are produced. A women can have a pair of ovaries, held by a membrane beside the uterus on each side of the lower abdomen. The ovary is responsible for producing the female reproductive cells or ova. During ovulation, a follicle can expels an egg under the stimulation of hormones. When an egg matures, it is released and passed into the fallopian tube towards the uterus. If the ovum is fertilized by the male reproductive cell or sperm, conception happens and pregnancy begins.

During ovulation or assisted reproduction, there are considerable changes in the size, volume, and appearance of the ovary, not only between the individuals but also based on the hormonal status.

An ultrasound imaging system can be used to assess ovarian parameters. The ultrasound imaging system irradiates an ultrasound signal, generated by a transducer of a probe, to the ovary and receives an echo signal. The echo signal is reflected from the ovary, thereby obtaining an image of a part inside the ovary. In particular, the ultrasound imaging system can used for medical purposes, such as internal observation of the ovary, diagnosis of damage in inside parts of the ovary, and so on.

The existing methods for quantifying the ovarian parameters can be operator dependent and time consuming. The quantified parameters may not be reproducible. An accurate delineation or segmentation of the ovarian region can aid a clinician in producing more accurate diagnosis.

Accordingly, the embodiments provide methods and systems for assessing an ovary of a subject for quantification of ovarian parameters by detection and localization of the ovary from ultrasound images. The embodiments include extracting a plurality of two dimensional (2D) radial slices by performing rotational slicing of a three dimensional (3D) scanned image of an ovary of a subject. The embodiments include performing an initial segmentation for determining coarse ovarian boundaries in each of the plurality of 2D radial slices. The embodiments include segmenting ovarian boundaries in each of the plurality of 2D radial slices based on the initial segmentation, for obtaining an ovarian region and a non-ovarian region. The embodiments include generating a 3D mesh structure of the ovary based on the ovarian boundaries in each of the plurality of 2D radial slices. The embodiments include quantifying the at least one ovarian parameter from the 3D mesh structure of the ovary for assessing the at least one ovarian parameter.

These and other aspects of the embodiments herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions, while indicating embodiments and numerous specific details thereof, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the embodiments herein without departing from the spirit thereof, and the embodiments herein include all such modifications.

According to the embodiments herein, methods and systems for providing aid to clinicians for assessment of ovarian parameters are provided.

According to the embodiments herein, there are technical advantages of providing aid to clinicians for assessment of the quantified ovarian parameters during In-Vitro Fertilization (IVF).

According to the embodiments herein, there are technical advantages of quantifying the ovarian parameters of a subject such as volume, diameter, morphology, blood flow, and other relevant ovarian parameters of the ovary, during stimulation cycle of the IVF; and tracking the quantified ovarian parameters throughout the stimulation cycle.

According to the embodiments herein, there are technical advantages of tracking the quantified ovarian parameters during predefined days of the stimulation cycle of the IVF cycle and comparing the quantified ovarian parameters of the subject with quantified ovarian parameters of other subjects during the stimulation cycle of the IVF.

According to the embodiments herein, there are technical advantages of classifying ovarian and non-ovarian structures, estimate ovarian reserve, detecting at least one ovarian disorder, classifying ovarian pathologies, and obtaining an enhanced visualization of the follicles in the ovary; based on quantified ovarian parameters.

According to the embodiments herein, there are technical advantages of retrieving quantified ovarian parameters of plurality of subjects and building nomograms based on a correlation between ovarian volume and ovarian reserve.

According to the embodiments herein, there are technical advantages of performing automated segmentation of ovarian region in three-dimensional ultrasound image of the ovary.

According to the embodiments herein, there are technical advantages of determining hormonal dosage for assisted reproduction based on the rate of change of quantified ovarian parameters.

According to the embodiments herein, there are technical advantages of performing staging of ovarian cancer based on the quantified ovarian parameters.

According to the embodiments herein, there are technical advantages of retrieving medical history of the subject (segmented ovarian region), for usage as a reference, for registering during current scanning of the ovary, administering follicle growth, tracking follicles for oocyte aspiration, analyzing rate of change of ovarian volume during stimulation cycle, predicting success of In-Vitro Fertilization (IVF), and so on.

This invention is illustrated in the accompanying drawings, through out which like reference letters indicate corresponding parts in the various figures. The embodiments herein will be better understood from the following description with reference to the drawings, in which:

FIG. 1 depicts anatomy of female reproductive system;

FIG. 2a and FIG. 2b are block diagrams illustrating structures of an apparatus for assessing an ovary, according to embodiments as disclosed herein;

FIG. 3a depicts an example tracking of variations in ovarian volume over a predefined period, according to embodiments as disclosed herein;

FIG. 3b depicts tracking of needle path for oocyte aspiration, according to embodiments as disclosed herein;

FIG. 4a depict examples of diagnosis of ovarian cancer, according to embodiments as disclosed herein;

FIG. 4b depict examples of diagnosis of Poly-Cystic Ovary Syndrome (PCOS) by quantifying follicles within the ovarian region, according to embodiments as disclosed herein;

FIG. 5 depicts an example enhanced visualization of the ovary, according to embodiments as disclosed herein;

FIG. 6 depicts example tracking of an ovary based on ovarian quantifications performed earlier, according to embodiments as disclosed herein;

FIGS. 7a-7c depict estimation of ovarian reserve from estimated ovarian volume based on nomograms, according to embodiments as disclosed herein;

FIG. 8 depicts classification of structures in an ultrasound image of an ovary, according to embodiments as disclosed herein;

FIGS. 9a and 9b depict example detection and removal of false detection of a structure outside the ovarian region, according to embodiments as disclosed herein;

FIGS. 9c and 9d depict example detection and removal of a leaking follicle based on segmented ovarian region, according to embodiments as disclosed herein; and

FIG. 10 is a flowchart depicting a method for accessing ovarian parameters from a scanned image of the ovary, according to embodiments as disclosed herein.

FIG. 11 is a block diagram illustrating an ultrasound diagnosis apparatus according to an exemplary embodiment.

FIGS. 12a, 12b, and 12c are diagrams respectively illustrating an ultrasound diagnosis apparatus according to an exemplary embodiment.

According to an aspect of an exemplary embodiment, an apparatus for assessing ovarian parameters is provided. The apparatus comprises an data obtaining device configured to obtain medical image data; and a processor configured to: extract a plurality of two dimensional (2D) radial slices by performing rotational slicing of a three dimensional (3D) scanned image of an ovary of a subject based on the obtained medical image data; perform an initial segmentation for determining coarse ovarian boundaries in each of the plurality of 2D radial slices; segment ovarian boundaries in each of the plurality of 2D radial slices based on the initial segmentation to obtain an ovarian region and a non-ovarian region; generate a 3D mesh structure of the ovary based on the ovarian boundaries in each of the plurality of 2D radial slices; and quantify the at least one ovarian parameter from the 3D mesh structure of the ovary.

The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

Certain exemplary embodiments are described in greater detail below with reference to the accompanying drawings.

In the following description, the same drawing reference numerals are used for the same elements even in different drawings. The matters defined in the description, such as detailed construction and elements, are provided to assist in a comprehensive understanding of exemplary embodiments. Thus, it is apparent that exemplary embodiments can be carried out without those specifically defined matters. Also, well-known functions or constructions are not described in detail since they would obscure exemplary embodiments with unnecessary detail.

Terms such as "part" and "portion" used herein denote those that may be embodied by software or hardware. According to exemplary embodiments, a plurality of parts or portions may be embodied by a single unit or element, or a single part or portion may include a plurality of elements.

In exemplary embodiments, an image may include any medical image acquired by various medical imaging apparatuses such as a magnetic resonance imaging (MRI) apparatus, a computed tomography (CT) apparatus, an ultrasound imaging apparatus, or an X-ray apparatus.

Also, in the present specification, an "object", which is a thing to be imaged, may include a human, an animal, or a part thereof. For example, an object may include a part of a human, that is, an organ or a tissue, or a phantom.

Throughout the specification, an ultrasound image refers to an image of an object processed based on ultrasound signals transmitted to the object and reflected therefrom.

The principal object of the embodiments herein is to disclose methods and systems for providing aid to clinicians for assessment of ovarian parameters.

Another object of the embodiments herein is to provide aid to clinicians for assessment of the quantified ovarian parameters during In-Vitro Fertilization (IVF).

Another object of the embodiments herein is to quantify the ovarian parameters of a subject such as volume, diameter, morphology, blood flow, and other relevant ovarian parameters of the ovary, during stimulation cycle of the IVF; and track the quantified ovarian parameters throughout the stimulation cycle.

Another object of the embodiments herein is to track the quantified ovarian parameters during predefined days of the stimulation cycle of the IVF cycle and compare the quantified ovarian parameters of the subject with quantified ovarian parameters of other subjects during the stimulation cycle of the IVF.

Another object of the embodiments herein is to classify ovarian and non-ovarian structures, estimate ovarian reserve, detect at least one ovarian disorder, classify ovarian pathologies, and obtain an enhanced visualization of the follicles in the ovary; based on quantified ovarian parameters.

Another object of the embodiments herein is to retrieve quantified ovarian parameters of plurality of subjects and build nomograms based on a correlation between ovarian volume and ovarian reserve.

Another object of the embodiments herein is to perform automated segmentation of ovarian region in three-dimensional ultrasound image of the ovary.

Another object of the embodiments herein is to determine hormonal dosage for assisted reproduction based on the rate of change of quantified ovarian parameters.

Another object of the embodiments herein is to perform staging of ovarian cancer based on the quantified ovarian parameters.

Another object of the embodiments herein is to retrieve medical history of the subject (segmented ovarian region), for usage as a reference, for registering during current scanning of the ovary, administering follicle growth, tracking follicles for oocyte aspiration, analyzing rate of change of ovarian volume during stimulation cycle, predicting success of In-Vitro Fertilization (IVF), and so on.

Embodiments herein disclose methods and systems for assessing an ovary of a subject for quantification of ovarian parameters by detection and segmentation of the ovary from a scanned image of an ovary. In an example, the scanned image can be obtained from a Trans-Vaginal Ultrasound (TVUS) scan of the ovary. The embodiments include analyzing the change of different ovarian parameters during In-Vitro Fertilization (IVF). The embodiments include obtaining a scanned image of the ovary of a subject, at predefined days within the IVF cycle. The scanned image can be used for determining the ovarian parameters of the subject such as size, volume, diameter, morphology, blood flow, and other relevant ovarian parameters. Any changes in the ovarian parameters detected using scanned images obtained at predefined days of the IVF cycle can be tracked. The embodiments include monitoring the changes of ovarian parameters in the IVF cycle. The embodiments include retrieving quantified ovarian parameters of plurality of subjects and build nomograms based on the correlation between ovarian volume and ovarian reserve. The embodiments include performing automated segmentation of ovarian region in three-dimensional ultrasound image of the ovary. The embodiments include determining hormonal dosage for assisted reproduction based on the rate of change of quantified ovarian parameters. The embodiments include performing staging of ovarian cancer based on the quantified ovarian parameters.

The embodiments include classifying structures within and outside the ovarian boundary based on location of the structures and the type of the structures. The structures can be classified as ovarian or non-ovarian based on the location of the structures in the ovarian region or the non-ovarian region. The type of structures can be classified as a follicle or a blood vessel.

The embodiments include tracking changes in ovarian and follicular volume over a predefined period. The embodiments include determining and tracking the path through which a needle can be guided for retrieving follicles for oocyte aspiration.

In an example embodiment herein, the ovarian reserve can be estimated based on the ovarian volume. In an example embodiment herein, ovarian disorders can be diagnosed based on blood flow around the ovarian boundaries. In an example embodiment herein, Poly-Cystic Ovary Syndrome (PCOS) can be diagnosed based on follicular volume quantification.

Referring now to the drawings, and more particularly to FIGS. 2 through 10, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments.

FIG. 2a is a block diagram illustrating a structure of an apparatus 200a for accessing an ovary, according to embodiments as disclosed herein. The apparatus 200a may comprise a data obtaining device 210 and a processor 220.

The data obtaining device 210 is configured to obtain medical image data. According to an exemplary embodiment, the data obtaining device 210 comprises a communication module, and receives the medical image data from another device via a network. According to another exemplary embodiment, the data obtaining device 210 comprises a device for scanning an object, and generates the medical image data by scanning the object. For example, the data obtaining device 210 comprises a probe for transmitting an ultrasound signal and receiving an echo signal. For example, the data obtaining device 210 comprises CT scanning device or MRI scanning device.

The processor 220 is configured to process data, and executed some functions based on at least one computer program instructions. The processor 220 may comprises one or more processor. The processor 220 is configured to extract a plurality of two dimensional (2D) radial slices by performing rotational slicing of a three dimensional (3D) scanned image of an ovary of a subject based on the obtained medical image data, perform an initial segmentation for determining coarse ovarian boundaries in each of the plurality of 2D radial slices, segment ovarian boundaries in each of the plurality of 2D radial slices based on the initial segmentation to obtain an ovarian region and a non-ovarian region, generate a 3D mesh structure of the ovary based on the ovarian boundaries in each of the plurality of 2D radial slices; and quantify the at least one ovarian parameter from the 3D mesh structure of the ovary.

FIG. 2b depicts various units of an apparatus 200b for accessing an ovary, according to embodiments as disclosed herein. As depicted in FIG. 2b, the apparatus 200b comprises a data obtaining device 210, a processor 220, and a display device. The processor 220 comprises a localization unit 201, a quantification unit 202, a classification unit 203, a tracking unit 204, a diagnosis unit 205. The localization unit 201 can detect an ovary from a 3D image, obtained from a TVUS scan of the ovary of a subject. The localization unit 201 can extract a plurality of 2D radial slices from the 3D scanned image of the ovary. In an embodiment herein, the extraction can be performed using a rotational based slicing approach. The radial slices can be separated from each other by a predefined degree, which is based on the number of radial slices to be extracted from the 3D scanned image. In an example, if 30 slices are extracted, then there is a separation of 6 degrees between the radial slices.

The localization unit 201 can perform segmentation on each of the plurality of 2D radial slices in order to determine initial ovarian boundaries in each of the plurality of radial slices. Learning based methods can be used for performing the segmentation in each of the plurality of radial slices.

The localization unit 201 can determine coarse ovarian boundaries in each of the plurality of 2D radial slices based on learning methods. In an embodiment, the deep learning can be used to generate initial segmentation.

The localization unit 201 can refine the coarse ovarian boundaries to determine (fine) ovarian boundaries in each of the plurality of radial slices. In an embodiment, the coarse ovarian boundaries in each of the plurality of 2D radial slices can be refined by an optimization method, in order to determine the (fine) ovarian boundaries. In an example, a level set based active contour deformation can be used. The level set functions can segment each of the plurality of 2D radial slices into an ovarian region (internal to the ovary) and a non-ovarian region (external to the ovary). The fine ovarian boundaries can be used for distinguishing the ovarian region and the non-ovarian region in each of the plurality of 2D radial slices. Henceforth, the fine ovarian boundaries will be referred ovarian boundaries.

The localization unit 201 can convert the plurality of 2D slices into a 3D mesh structure. In an embodiment, the 3D mesh structure can be generated by performing spherical parameterization based on the segmented ovarian boundaries in each of the 2D radial slices. The spherical parameterization can be performed along a latitude on the surface of the ovary (herein after referred to as a first direction) and longitude along the ovarian boundaries in each of the 2D radial slices (herein after referred to as a second direction). The parameter value along the first direction and the second direction can be based on a normalized distance along the first direction and the second direction.

The spherical parameterization can be further used for generating a mesh. A mesh can be represented by a plurality of triangles. The plurality of triangles can be used for constructing the 3D mesh structure by connecting adjacent nodes along the first direction and the second direction.

The quantification unit 202 can quantify the ovarian volume by estimating the surface of the ovary from the 3D mesh structure. The quantification unit 202 can quantify structures, such as follicles, blood vessels, in the ovarian and the non-ovarian region. The quantification unit 202 can estimate the ovarian reserve based on the quantified ovarian volume. The quantification unit 202 can quantify the blood flow in the ovarian and non-ovarian regions. The quantification unit 202 can quantify the morphology of the 3D ovarian surface.

The classification unit 203 can utilize the results of the segmentation to classify structures in the scanned image based on the segmentation of the ovary. If the whole structure is within the ovarian region, i.e., within the ovarian boundary, then the structure can be classified as ovarian. If a part of the structure or the whole structure is in the non-ovarian region, i.e., outside the ovarian boundary, then the structure can be classified as non-ovarian. The non-ovarian structures can be removed.

In an example, consider that a follicle is detected in the non-ovarian region. In another example, a leaked follicle can be detected, wherein a part of the follicle can be in the ovarian region and a part of the follicle can be in the non-ovarian region. In both scenarios, the follicle in the non-ovarian region and the non-ovarian region can be removed by the classification unit 203.

The classification unit 203 can classify structures, within or outside the ovarian region. The structures can be classified based on its type, such as blood vessels, follicles, and so on.

The classification unit 203 can classify the structures in the scanned image to determine staging of ovarian cancer

The classification unit 203 can classify the blood flow within or outside the ovarian region. The blood flow can be used to determining staging of ovarian cancer.

The tracking unit 204 can track variations, if any, in the ovarian volume over a period of time. In an example, the tracking unit 204 can quantify the ovarian volume of a subject each day for a predefined number of days. Ovarian quantifications such as ovarian volume and follicular volume, quantified each day, can be stored in a database for tracking. In an example, the ovarian volume of the subject quantified at a certain day can be retrieved at a later date.

The tracking unit 204 can track the path of a needle used to retrieve oocytes from specific follicles in a particular location of the ovary. The oocytes can be obtained by choosing follicles in the ovary with a larger volume, which have been localized or segmented by the localization unit 201.

The diagnosis unit 205 can perform feature extraction of the structures, detected in the ovary, based on quantification of the structures. In an example, the feature extraction can be performed to diagnose disorders such as Poly-Cystic Ovary Syndrome (PCOS). In an example, the diagnosis unit 205 can diagnose ovarian cancer based on blood flow within and outside the ovarian boundaries.

The display device 206 can display the initial and refined ovarian boundaries in each of the 2D radial slices, and the 3D mesh structure of the ovary. The display device 206 can provide an enhanced visualization of the ovary, wherein the ovarian boundary and the follicles within the boundary are segmented and defined. The display device 206 can display the validity of the diagnosis of the diagnosis unit 205 by displaying the ovarian disorders. The display device 206 provides visualization of the ovarian and non-ovarian structures, and their type. The display device 206 provides visualization of detection of false positives and leaked follicles their subsequent removal.

FIG. 2a and FIG. 2b show exemplary units of the

apparatus

200a and 200b, but it is to be understood that other embodiments are not limited thereon. In other embodiments, the

apparatus

200a and 200b may include less or more number of units. Further, the labels or names of the units are used only for illustrative purpose and does not limit the scope of the invention. One or more units can be combined together to perform same or substantially similar function in the

apparatus

200a and 200b. In this specification, an apparatus 200 indicates the

apparatus

200a and 200b.

FIG. 3a depicts an example tracking of variations in ovarian volume over a predefined period, according to embodiments as disclosed herein. When a subject/patient visits a qualified person such as a technician, gynecologist, and so on, the ovary of the subject can be scanned using the apparatus 200. The technician or the gynecologist can be co-located with the subject or can be remotely located from the subject. Consider that the ovary of the subject has been scanned previously for two days, viz., day 0 and day 1. During each of those days, the ovary/follicles can be segmented and the volume of the ovary and the follicles can be quantified. On day 2, the apparatus 200can perform quantification of the ovary and load the quantifications of the ovarian volume and the follicles on day 0 and day 1 from a database. The apparatus 200can perform longitudinal tracking of changes in the volume of the ovary and the follicles. Ovary and follicle tracking can be performed, as the relative positions of the follicles with respect to ovary do not change. Effective ovarian tracking can improve follicular tracking during In Vitro Fertilization (IVF) cycles.

FIG. 3b depicts tracking of needle path for oocyte aspiration, according to embodiments as disclosed herein. For oocyte aspiration a follicle with a greater volume can be selected. The embodiments include guiding the needle though a path, such that the needle reaches the follicle with the greater volume.

Changes in the volume of the ovary and the volume of the follicles, over a period of days, can be tracked by quantifying the volume of the ovary and the volume of the follicles each day. The ovary can be tracked by superimposition of a scan of the ovary, performed on a previous day, with a scan of the ovary performed on the present day. The superimposition involves mapping of the follicles in the scans. Based on the superimposition, the rate of change in the volume of the ovary and the volume of the follicles can be determined. A follicle which a greater positive rate of change can be selected for retrieval.

FIG. 4a depicts an example diagnosis of ovarian cancer, according to embodiments as disclosed herein. The symptoms of occurrence of ovarian cancer can be detected based on the segmented ovarian boundary and blood flow across the segmented ovarian boundary.

FIG. 4b depicts an example diagnosis of PCOS by quantifying follicles within the ovarian region, according to embodiments as disclosed herein. PCOS is a situation where, a subject can have infrequent or prolonged menstrual periods. The ovaries may develop numerous small follicles and fail to regularly release eggs. This can be one of the major causes of infertility. PCOS classification is clinically important for early prediction of IVF outcomes and treatment methods. The localization of the ovarian boundary and quantification of the ovarian volume and the follicles within the ovarian volume can be used as a clinical biomarker for PCOS classification. A comparison between a normal ovary and an ovary with PCOS is depicted in FIG. 4b, wherein the ovary with PCOS is having numerous small follicles.

FIG. 5 depicts an example enhanced visualization of the ovary, according to embodiments as disclosed herein. As depicted in FIG. 5, the ovary and the follicles inside the ovary are segmented and quantified. The surface of the ovary can be estimated and each follicle inside the surface of the ovary can be numbered for tracking. This allows follicle tracking. In an example, the follicles are numbered in the range 2-8. This provides an enhanced visualization of the ovary and the follicles inside the ovary.

FIG. 6 depicts example tracking of an ovary based on ovarian quantifications performed earlier, according to embodiments as disclosed herein. Consider that a quantification of the ovarian volume was performed on a previous day. The quantification is performed by segmenting the ovary and the follicles inside the ovary, as depicted in 601. The volume of the ovary and the volume of the follicles inside the ovary are determined, as depicted in 603.

On the present day, the ovarian volume is again quantified by segmenting the ovary and the follicles, as depicted in 602. The follicles detected in the ovary and are labeled 1-6. The volume of the ovary and the volume of the follicles are determined, as depicted in 604. As depicted in 604, it is detected that the volume of the follicles has increased. A structure 605, in the non-ovarian region, is detected on both the previous and the present days.

The embodiments can retrieve ovarian quantifications, quantified on the previous day, such as the segmented ovarian region, the ovarian volume, and the volume of the follicles, from a database. These ovarian quantifications can be used as a reference for registering the follicles and administering the follicle growth on the present day, as depicted in 606. The registration allows tracking the growth of the follicular shape for oocyte aspiration, and analyzing the rate of change of the ovarian volume during the stimulation cycle. This allows predicting the success of the IVF process.

A comparison of follicular volume between the present day and the previous day is depicted in 607. The ovarian boundary, ovarian volume and the follicular volume on the present day are depicted in 608.

FIGS. 7a-7c depict estimation of ovarian reserve from estimated ovarian volume based on nomograms, according to embodiments as disclosed herein. The ovarian reserve can be used for determining the capacity of the ovary to provide egg cells, which are capable of fertilization resulting in a healthy and successful pregnancy. At advanced maternal ages, the number of eggs that can be successfully fertilized for pregnancy can decline.

As depicted in FIG. 7a, the 3D mesh structure of the ovary can be used for estimating the surface of the ovary and determining the ovarian volume.

FIG. 7b is an example graph depicting a relation between ovarian volume and reproductive age of subjects of chronological age 35. The ovarian parameters can be quantified at predefined days during the IVF cycle. Based on the quantified ovarian parameters at the predefined days, for a plurality of subjects, the embodiments include generating and displaying nomograms. The nomograms can be used to predict the ovarian reserve based on the ovarian volume.

FIG. 7c is a nomogram indicating the relation between chronological age of subjects, ovarian reserve and reproductive age.

The estimation of the ovarian reserve can help in early prediction of successful IVF procedures and help in reducing cost of going through multiple IVF cycles and hormonal impact. The estimation of the ovarian reserve can be used as an indicator for initiating oocyte cryopreservation.

FIG. 8 depicts classification of structures in an ultrasound image of an ovary, according to embodiments as disclosed herein. As depicted in FIG. 8, different structures (tissues) in the ovarian region and the non-ovarian region can be classified through a pelvic ultrasound scan. In an example, structures viz., the follicles '1', the ovarian boundary '2', and the blood vessels '3' are detected and classified based on their type.

FIGS. 9a and 9b depict example detection and removal of false detection of a structure outside the ovarian region, according to embodiments as disclosed herein. The embodiments can classify the detected structures as ovarian or non-ovarian based on their location (ovarian or non-ovarian region). A Region of Interest (ROI) can be automatically selected. As depicted in FIG. 9a, a structure 'A' (follicle) is detected within the ROI, which is outside the ovarian boundary, i.e., in the non-ovarian region. As depicted in FIG. 9b, the embodiments can remove the structure 'A'.

FIGS. 9c and 9d depict example detection and removal of a leaking follicle based on segmented ovarian region, according to embodiments as disclosed herein. As depicted in FIG. 9c, 'A' represents the ovarian boundary. The embodiments can detect a leaking follicle 'B', and classify 'B' as a non-ovarian structure. Subsequently, as depicted in FIG. 9d, the embodiments can remove the leaking follicle 'B'.

Automating the quantification of ovarian volume can speed up the clinical workflow, reduce operator bias, and aid in harnessing the full potential of the ultrasound in clinical practice. Embodiments herein can delineate the ovarian boundary from the ultrasound volumes.

FIG. 10 is a flowchart 1000 depicting a method for accessing ovarian parameters from a scanned image of the ovary, according to embodiments as disclosed herein. At step 1001, the method includes extracting a plurality of radial slices by performing rotational slicing of a scanned image of an ovary of a subject. The scanned image of the ovary can be obtained from a TVUS scan of the ovary of a subject. The extraction can be performed using a rotational based slicing approach. The radial slices can be separated from each other by a predefined degree, which is based on the number of radial slices to be extracted from the scanned image.

At step 1002, the method includes generating an initial ovarian boundary in each of the radial slices. Learning based technique can be used to generate the initial ovarian boundary. In an embodiment, the deep learned energy map can be generated using U-Net architecture.

At step 1003, the method includes segmenting ovarian boundaries in each of the plurality of 2D radial slices. Initially coarse ovarian boundaries can be determined. The coarse ovarian boundaries can be refined in order to determine (fine) ovarian boundary in each of the plurality of 2D radial slices. The coarse ovarian boundaries can be refined by minimizing a cost function .In an embodiment, level set based active contour can be used to minimize the cost function to obtain the refined (fine) ovarian boundaries. The level set functions partitions each of the plurality of radial slices into an external region and an internal region. The internal region is the ovarian region and the external region is the non-ovarian region.

At step 1004, the method includes generating the 3D mesh structure of the ovary based on the segmented ovarian boundaries in each of the plurality of 2D radial slices. The 3D mesh structure can be generated by performing spherical parameterization based on the segmented ovarian boundaries in each of the 2D radial slices. The spherical parameterization can be performed along latitude on the surface of the ovary and along the ovarian boundaries in each of the 2D radial slices. The spherical parameterization can be used for generating a plurality of triangles. The plurality of triangles can be used for constructing the 3D mesh structure by connecting adjacent nodes along the latitude on the surface of the ovary and along the ovarian boundaries in each of the 2D radial slices.

At step 1005, the method includes quantifying the volume of the ovary from 3D mesh structure of the ovary. The quantification of the ovarian volume can be performed by estimating the surface of the ovary from the 3D mesh structure.

The various actions in method 1100 may be performed in the order presented, in a different order or simultaneously. Further, in some embodiments, some actions listed in FIG. 10 may be omitted.

The embodiments disclosed herein can be implemented through at least one software program running on at least one hardware device and performing network management functions to control the network elements. The network elements shown in FIG. 2 include blocks which can be at least one of a hardware device, or a combination of hardware device and software module.

The embodiments disclosed herein describe methods and systems for assessing an ovary of a subject for quantification of ovarian parameters by detection and segmentation of the ovary from ultrasound images. Therefore, it is understood that the scope of the protection is extended to such a program and in addition to a computer readable means having a message therein, such computer readable storage means contain program code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The method is implemented in a preferred embodiment through or together with a software program written in e.g. Very high speed integrated circuit Hardware Description Language (VHDL) another programming language, or implemented by one or more VHDL or several software modules being executed on at least one hardware device. The hardware device can be any kind of portable device that can be programmed. The device may also include means which could be e.g. hardware means like e.g. an ASIC, or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software modules located therein. The method embodiments described herein could be implemented partly in hardware and partly in software. Alternatively, the invention may be implemented on different hardware devices, e.g. using a plurality of CPUs.

FIG. 11 is a block diagram illustrating a configuration of an ultrasound diagnosis apparatus 100, i.e., a diagnostic apparatus, according to an exemplary embodiment. The apparatus 200 may embodied as an ultrasound diagnosis apparatus 100.

Referring to FIG. 11, the ultrasound diagnosis apparatus 100 may include a probe 20, an ultrasound transceiver 110, a controller 120, an image processor 130, one or more displays 140, a storage 150, e.g., a memory, a communicator 160, i.e., a communication device or an interface, and an input interface 170.

The ultrasound diagnosis apparatus 100 may be of a cart-type or a portable-type ultrasound diagnosis apparatus, that is portable, moveable, mobile, or hand-held. Examples of the portable-type ultrasound diagnosis apparatus 100 may include a smart phone, a laptop computer, a personal digital assistant (PDA), and a tablet personal computer (PC), each of which may include a probe and a software application, but embodiments are not limited thereto.

The probe 20 may include a plurality of transducers. The plurality of transducers may transmit ultrasound signals to an object 10 in response to transmitting signals received by the probe 20, from a transmitter 113. The plurality of transducers may receive ultrasound signals reflected from the object 10 to generate reception signals. In addition, the probe 20 and the ultrasound diagnosis apparatus 100 may be formed in one body (e.g., disposed in a single housing), or the probe 20 and the ultrasound diagnosis apparatus 100 may be formed separately (e.g., disposed separately in separate housings) but linked wirelessly or via wires. In addition, the ultrasound diagnosis apparatus 100 may include one or more probes 20 according to embodiments.

The controller 120 may control the transmitter 113 for the transmitter 113 to generate transmitting signals to be applied to each of the plurality of transducers based on a position and a focal point of the plurality of transducers included in the probe 20.

The controller 120 may control the ultrasound receiver 115 to generate ultrasound data by converting reception signals received from the probe 20 from analogue to digital signals and summing the reception signals converted into digital form, based on a position and a focal point of the plurality of transducers.

The image processor 130 may generate an ultrasound image by using ultrasound data generated from the ultrasound receiver 115.

The display 140 may display a generated ultrasound image and various pieces of information processed by the ultrasound diagnosis apparatus 100. The ultrasound diagnosis apparatus 100 may include two or more displays 140 according to the present exemplary embodiment. The display 140 may include a touch screen in combination with a touch panel.

The controller 120 may control the operations of the ultrasound diagnosis apparatus 100 and flow of signals between the internal elements of the ultrasound diagnosis apparatus 100. The controller 120 may include a memory for storing a program or data to perform functions of the ultrasound diagnosis apparatus 100 and a processor and/or a microprocessor (not shown) for processing the program or data. For example, the controller 120 may control the operation of the ultrasound diagnosis apparatus 100 by receiving a control signal from the input interface 170 or an external apparatus.

The ultrasound diagnosis apparatus 100 may include the communicator 160 and may be connected to external apparatuses, for example, servers, medical apparatuses, and portable devices such as smart phones, tablet personal computers (PCs), wearable devices, etc., via the communicator 160.

The communicator 160 may include at least one element capable of communicating with the external apparatuses. For example, the communicator 160 may include at least one among a short-range communication module, a wired communication module, and a wireless communication module.

The communicator 160 may receive a control signal and data from an external apparatus and transmit the received control signal to the controller 120 so that the controller 120 may control the ultrasound diagnosis apparatus 100 in response to the received control signal.

The controller 120 may transmit a control signal to the external apparatus via the communicator 160 so that the external apparatus may be controlled in response to the control signal of the controller 120.

For example, the external apparatus connected to the ultrasound diagnosis apparatus 100 may process the data of the external apparatus in response to the control signal of the controller 120 received via the communicator 160.

A program for controlling the ultrasound diagnosis apparatus 100 may be installed in the external apparatus. The program may include command languages to perform part of operation of the controller 120 or the entire operation of the controller 120.

The program may be pre-installed in the external apparatus or may be installed by a user of the external apparatus by downloading the program from a server that provides applications. The server that provides applications may include a recording medium where the program is stored.

The storage 150 may store various data or programs for driving and controlling the ultrasound diagnosis apparatus 100, input and/or output ultrasound data, ultrasound images, applications, etc.

The input interface 170 may receive a user's input to control the ultrasound diagnosis apparatus 100 and may include a keyboard, button, keypad, mouse, trackball, jog switch, knob, a touchpad, a touch screen, a microphone, a motion input means, a biometrics input means, etc. For example, the user's input may include inputs for manipulating buttons, keypads, mice, trackballs, jog switches, or knobs, inputs for touching a touchpad or a touch screen, a voice input, a motion input, and a bioinformation input, for example, iris recognition or fingerprint recognition, but an exemplary embodiment is not limited thereto.

Referring to FIGS 2a, 2b, and 11, according to an exemplary embodiment, the data obtaining device 210 correspond to the probe 20 and ultrasound transceiver 110. According to another exemplary embodiment, the data obtaining device 210 corresponds to the communicator 160. According to an exemplary embodiment, the processor 220 corresponds to the controller 120 and the image processor 130. According to an exemplary embodiment, the display device 206 corresponds to the display 140.

An example of the ultrasound diagnosis apparatus 100 according to the present exemplary embodiment is described below with reference to FIGS. 12a, 12b, and 12c.

FIGS. 12a, 12b, and 12c are diagrams illustrating ultrasound diagnosis apparatus according to an exemplary embodiment.

Referring to FIGS. 12a and 12b, the ultrasound diagnosis apparatus 100 may include a main display 121 and a sub-display 122. At least one among the main display 121 and the sub-display 122 may include a touch screen. The main display 121 and the sub-display 122 may display ultrasound images and/or various information processed by the ultrasound diagnosis apparatus 100. The main display 121 and the sub-display 122 may provide graphical user interfaces (GUI), thereby receiving user's inputs of data to control the ultrasound diagnosis apparatus 100. For example, the main display 121 may display an ultrasound image and the sub-display 122 may display a control panel to control display of the ultrasound image as a GUI. The sub-display 122 may receive an input of data to control the display of an image through the control panel displayed as a GUI. The ultrasound diagnosis apparatus 100 may control the display of the ultrasound image on the main display 121 by using the input control data.

Referring to FIG. 12b, the ultrasound diagnosis apparatus 100 may include a control panel 165. The control panel 165 may include buttons, trackballs, jog switches, or knobs, and may receive data to control the ultrasound diagnosis apparatus 100 from the user. For example, the control panel 165 may include a time gain compensation (TGC) button 171 and a freeze button 172. The TGC button 171 is to set a TGC value for each depth of an ultrasound image. Also, when an input of the freeze button 172 is detected during scanning an ultrasound image, the ultrasound diagnosis apparatus 100 may keep displaying a frame image at that time point.

The buttons, trackballs, jog switches, and knobs included in the control panel 165 may be provided as a GUI to the main display 121 or the sub-display 122.

Referring to FIG. 12c, the ultrasound diagnosis apparatus 100 may include a portable device. An example of the portable ultrasound diagnosis apparatus 100 may include, for example, smart phones including probes and applications, laptop computers, personal digital assistants (PDAs), or tablet PCs, but an exemplary embodiment is not limited thereto.

The ultrasound diagnosis apparatus 100 may include the probe 20 and a main body 40. The probe 20 may be connected to one side of the main body 40 by wire or wirelessly. The main body 40 may include a touch screen 145. The touch screen 145 may display an ultrasound image, various pieces of information processed by the ultrasound diagnosis apparatus 100, and a GUI.

The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the spirit and scope of the embodiments as described herein.

Claims

An apparatus for assessing ovarian parameters comprising:

an data obtaining device configured to obtain medical image data; and

a processor configured to:

extract a plurality of two dimensional (2D) radial slices by performing rotational slicing of a three dimensional (3D) scanned image of an ovary of a subject based on the obtained medical image data;

perform an initial segmentation for determining coarse ovarian boundaries in each of the plurality of 2D radial slices;

segment ovarian boundaries in each of the plurality of 2D radial slices based on the initial segmentation to obtain an ovarian region and a non-ovarian region;

generate a 3D mesh structure of the ovary based on the ovarian boundaries in each of the plurality of 2D radial slices; and

quantify the at least one ovarian parameter from the 3D mesh structure of the ovary.
The apparatus of claim 1, wherein the at least one ovarian parameter is size of the ovary, ovarian volume, morphology of the ovary, and blood flow in the ovary.
The apparatus of claim 2, wherein the processor is further configured to estimate an ovarian reserve based on the ovarian volume.
The apparatus of claim 1, wherein the processor is further configured to generate a nomogram based on comparison of the at least one ovarian parameter of the subject with at least one ovarian parameter of a plurality of subjects.
The apparatus of claim 1, further comprising a diagnosis device configured to diagnose ovarian cancer and a stage of the ovarian cancer based on blood flow in the ovarian region and the non-ovarian region.
The apparatus of claim 1, wherein the processor is further configured to diagnose Poly-Cystic Ovary Syndrome (PCOS) based on the at least one ovarian parameter.
The apparatus of claim 1, wherein the processor is further configured to track a change in the at least one ovarian parameter.
The apparatus of claim 7, wherein the processor is further configured to determine hormonal dosage for stimulation of the ovary during an In-Vitro Fertilization (IVF) cycle, based on the changes in the at least one ovarian parameter.
The apparatus of claim 1, wherein the processor is further configured to classify a structure based on at least one of location of the structure and type of the structure, wherein the location is one of the ovarian region and the non-ovarian region, wherein the type is one of follicles, cysts, adnexal masses, adnexa and blood vessels.
The apparatus of claim 1, wherein the 3D mesh structure of the ovary is generated by:

performing a spherical parameterization along latitude on the surface of the ovary and along the ovarian boundary in each of the plurality of 2D radial slices, wherein parameter value along the latitude on the surface of the ovary and along the ovarian boundary is based on normalized distance along the latitude on the surface of the ovary and along the ovarian boundary;

generating a plurality of triangles based on the spherical parameterization, wherein number of discretized nodes along the latitude on the surface of the ovary and along the ovarian boundary is based on number of 2D radial slices and number of sampled points on the ovarian boundaries in each of the plurality of 2D slices; and

constructing the 3D mesh structure through the plurality of triangles by connecting adjacent nodes along the longitudes on the surface of the ovary and latitudes along the ovarian boundary in each of the plurality of 2D radial slices.
The apparatus of claim 1, wherein the processor is further configured to:

obtain an ovarian volume and a follicular volume quantified on at least one previous day, wherein the ovarian volume and the follicular volume is stored in a database;

quantify an ovarian volume and a follicular volume on a present day; and

superimpose the ovarian volume and the follicular volume quantified on the present day on the ovarian volume and the follicular volume quantified on the at least one previous day.
The apparatus of claim 11, wherein the processor is further configured to:

determine follicles greater than a predefined volume among the follicles of the ovary, wherein the follicles greater than the predefined volume are determined based on the superimposition; and

determine a path to be followed by a needle probe for retrieving the follicles during oocyte aspiration.
A method for assessing at least one ovarian parameter, the method comprising:

extracting, by a processor, a plurality of two dimensional (2D) radial slices by performing rotational slicing of a three dimensional (3D) scanned image of an ovary of a subject;

performing, by the processor, an initial segmentation for determining coarse ovarian boundaries in each of the plurality of 2D radial slices;

segmenting, by the processor, ovarian boundaries in each of the plurality of 2D radial slices based on the initial segmentation to obtain an ovarian region and a non-ovarian region;

generating, by the processor, a 3D mesh structure of the ovary based on the ovarian boundaries in each of the plurality of 2D radial slices; and

quantifying, by the processor, the at least one ovarian parameter from the 3D mesh structure of the ovary.
The method of claim 13, wherein the at least one ovarian parameter is size of the ovary, ovarian volume, morphology of the ovary, and blood flow in the ovary.
The method of claim 14, wherein the method further comprises estimating, by the processor, an ovarian reserve based on the ovarian volume.