TWI270363B - Systems and methods for automated ventilator waveform recognition and measure based on ontologies - Google Patents
Systems and methods for automated ventilator waveform recognition and measure based on ontologies Download PDFInfo
- Publication number
- TWI270363B TWI270363B TW94145529A TW94145529A TWI270363B TW I270363 B TWI270363 B TW I270363B TW 94145529 A TW94145529 A TW 94145529A TW 94145529 A TW94145529 A TW 94145529A TW I270363 B TWI270363 B TW I270363B
- Authority
- TW
- Taiwan
- Prior art keywords
- point
- respiratory
- breathing
- waveform
- slope
- Prior art date
Links
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Air Conditioning Control Device (AREA)
- Control Of Positive-Displacement Air Blowers (AREA)
Abstract
Description
1270363 策 九、發明說明: 【發明所屬之技術領域】 此發明是一種醫療之電腦輔助技術,特別是一種基於知識本體之自動 化呼吸波形判讀暨量測方法及系統。 【先前技術】 目如呼吸器監控畫面上所顯示之壓力·時間、流速_時間、容積_時間及 谷積-壓力圖已成為呼吸治療診斷上一種不可或缺的資訊。目前胸腔專科醫 師都是透過聽診搭配呼健制_呼吸參數值,進行病患呼吸相關病症 ⑩ 的診斷,但這對於病患呼吸病症的嚴重程度(相對於正常狀況程度)不易有準 * 確疋畺數值的判定,故本專利提出一基於知識本體之呼吸波形判讀、量測 , 機制與系統,可判讀一完整週期之呼吸波型的呼吸氣道壓力、呼吸時間與 呼吸波形斜率可能性分佈,並量測出與正常波形的差異程度,以將其偏離 正常之嚴重程度數值化,提供醫師做更準確之判斷。 【發明内容】 由於呼吸病患為數者眾,目前呼吸器監控晝面上所顯示之壓力-時間、 流速-時間、容積-時間及容積_壓力圖已成為呼吸治療診斷上一種不可或缺 魯 的資成。目如胸腔專科醫師都是透過聽診搭配呼吸器量測到的呼吸參數 值,進行病患呼吸相關病症的診斷,但這對於病患呼吸病症的嚴重程度(相 對於正常狀況程度)不易有準確定量數值的判定。本發明的目的在於解決上 it問通以模糊邏輯為基礎,進而建構一套基於知識本體之智慧型呼吸波 幵乂判項機制與系統,以節省醫療資源。藉由建立呼吸氣道壓力可能性分佈 ' 及呼吸時間可能性分佈,再透過模糊波形斜率建構演算法(Fuzzy Wavefonn ' SlQpe e()n_etic)n Algorithm: FWSCA)來推論出呼吸氣道壓力對呼吸時間 的呼吸波形斜率可能性分佈,以進一步計算出輸入呼吸波形屬於正常波形 的可能性。 依據上述目的,本發明實施例揭露一種基於知識本體之自動化呼吸波 Chenfs Docket N〇.:ACI94022 s Docket No:0213-A40656-TW/Final/Jonah/20051117 5 1270363 -形·暨制綠,被—處科元職行,其枝包括··提仙應於實際 呼讀形中的第—職狀第—呼吸時_與第-呼吸氣道壓力,以及相 應於實際呼吸波形中的第二特徵點之第二啤吸時間點與第二呼吸氣道壓 力二依據第-呼吸時間點、第—呼吸氣道壓力、第二呼吸時間點與第二呼 吸氣道壓力,β十算第-特徵點與第二特徵點間之呼吸波形斜率;依據呼吸 波形斜率之可紐分佈資訊,取得減於呼吸波形斜率之呼吸波形斜率隸 屬程度;以及顯示呼吸波形斜率隸屬程度。 本發明實施例另揭露-觀腦可讀取贿媒體,用以儲存電腦程式, • _腦程式用以載人至電齡統巾並且使得該電齡統執行如上所述之基 於知識本體之自動化呼吸波形判讀暨量測方法。 本發明實施例更揭鉻一種基於知識本體之自動化呼吸波形判讀暨量 測系統,包括一個輸出單元與一個處理單元。處理單元耦接於輸出單元, 用以提供相應於實際呼吸波形中的第一特徵點之第一呼吸時間點與第一呼 吸氣道壓力,以及相應於實際呼吸波形中的第二特徵點之第二呼吸時間點 與第二呼吸氣道壓力;依據第一呼吸時間點、第一呼吸氣道壓力、第二呼 吸時間點與第二呼吸氣道壓力,計算第一特徵點與上述第二特徵點間之呼 鲁 吸波形斜率;依據呼吸波形斜率之可能性分佈資訊,取得相應於呼吸波形 斜率之呼吸波形斜率隸屬程度;以及透過輸出單元顯示呼吸波形斜率隸屬 程度。 其中,呼吸波形斜率之可能性分佈資訊係使用訓練方法而得。 【實施方式】 • 第1圖係表示依據本發明實施例之基於知識本體之自動化呼吸波形判 項暨篁測糸統10之硬體架構圖’包括處理單元11、記憶體12、儲存裝置 13、輸出裝置14、輸入裝置15、通訊裝置16,並使用匯流排17將其連結 在一起。除此之外,熟習此技藝人士也可將此系統實施於其他電腦系統樣 態(configuration)上,例如,手持式設備(hand-held devices)、多處理器系統、1270363 策 九, invention description: [Technical field of invention] This invention is a computer-aided technology for medical treatment, in particular, an automatic breathing waveform interpretation and measurement method and system based on knowledge ontology. [Prior Art] The pressure, time, flow rate_time, volume_time, and grain-pressure map displayed on the respirator monitoring screen have become an indispensable information for the diagnosis of respiratory therapy. At present, the thoracic specialists are diagnosed with the patient's respiratory related symptoms by auscultation with the respiratory parameter _ breathing parameter value, but this is not easy to be accurate for the severity of the patient's respiratory disease (relative to the normal condition). The determination of the value of 畺, so this patent proposes a respiratory waveform interpretation, measurement, mechanism and system based on the knowledge ontology, which can interpret the respiratory airway pressure, respiratory time and respiratory waveform slope probability distribution of a complete cycle of respiratory waveforms, and The degree of difference from the normal waveform is measured to quantify its deviation from normal severity, providing the physician with a more accurate judgment. SUMMARY OF THE INVENTION Because of the large number of respiratory patients, the current pressure-time, flow-time, volume-time and volume_pressure maps displayed on the respirator monitoring surface have become an indispensable part of the diagnosis of respiratory therapy. Capital. For example, the thoracic specialists diagnose the respiratory-related conditions of the patient by auscultation with the values of the respiratory parameters measured by the respirator, but this is not easy to accurately quantify the severity of the respiratory condition of the patient (relative to the normal condition). The determination of the value. The object of the present invention is to solve the problem that the upper it is based on fuzzy logic, and then construct a set of intelligent respiratory wave sputum judgment mechanism and system based on knowledge ontology to save medical resources. By establishing a respiratory airway pressure probability distribution' and a breathing time probability distribution, and then using the fuzzy waveform slope construction algorithm (Fuzzy Wavefonn 'SlQpe e()n_etic)n Algorithm: FWSCA) to infer the respiratory airway pressure versus breathing time The respiratory waveform slope probability distribution is further calculated to further calculate the likelihood that the input respiratory waveform is a normal waveform. According to the above object, an embodiment of the present invention discloses an automatic breathing wave based on the knowledge ontology Chenfs Docket N〇.: ACI94022 s Docket No:0213-A40656-TW/Final/Jonah/20051117 5 1270363 - Shape · cum green, by - In the department of the department, the branch includes: the first position in the actual call form - the breathing time _ and the first - respiratory airway pressure, and the second characteristic point corresponding to the actual respiratory waveform The second beer suction time point and the second respiratory airway pressure 2 are based on the first-breathing time point, the first-breathing airway pressure, the second breathing time point and the second respiratory airway pressure, and the beta-characteristic-characteristic point and the second characteristic point The slope of the respiratory waveform; the degree of membership of the respiratory waveform minus the slope of the respiratory waveform is obtained based on the information of the slope of the respiratory waveform; and the degree of membership of the respiratory waveform is displayed. According to another embodiment of the present invention, a brain-reading bribery medium is used for storing a computer program, and a brain program is used to carry a person to an electric age towel and enable the electronic age system to perform the automation based on the knowledge ontology as described above. Respiratory waveform interpretation and measurement method. The embodiment of the invention further discloses an automatic breathing waveform interpretation and measurement system based on the knowledge ontology, comprising an output unit and a processing unit. The processing unit is coupled to the output unit for providing a first breathing time point corresponding to the first feature point in the actual respiratory waveform and the first respiratory airway pressure, and a second corresponding to the second characteristic point in the actual respiratory waveform a breathing time point and a second breathing airway pressure; calculating a Hulu between the first feature point and the second feature point according to the first breathing time point, the first breathing airway pressure, the second breathing time point, and the second respiratory airway pressure The slope of the waveform is absorbed; according to the probability distribution information of the slope of the respiratory waveform, the degree of membership of the respiratory waveform corresponding to the slope of the respiratory waveform is obtained; and the degree of membership of the respiratory waveform is displayed through the output unit. Among them, the probability distribution information of the slope of the respiratory waveform is obtained by using a training method. [Embodiment] FIG. 1 is a diagram showing a hardware structure diagram of an automatic breathing waveform judgment and measurement system based on an ontology according to an embodiment of the present invention, including a processing unit 11, a memory 12, a storage device 13, The output device 14, the input device 15, and the communication device 16 are connected together using the bus bar 17. In addition, those skilled in the art can implement this system on other computer system configurations, such as hand-held devices, multi-processor systems,
Client’s Docket No.:ACI94022 TT^ Docket No:0213-A40656-TW/Final/Jonah/20051117 ⑧ 1270363 Λ ·以微處理器為基礎或可程式化之消費性電子產品(miCr〇process〇r_baSed οι* consumer electronics)、網路電腦、迷你電腦、大型主機以及 類似之設備。處理單元11可包含一單一中央處理單元(central_pr〇cessing_ cpu)或者是關連於平行運算環境__ processing 之多個平5 行處理單元。3己憶體12包含唯讀記憶體(1^〇11以11^111〇1^;1^〇]^)、快閃記 憶體(£^11〇]\4)以及/或動態存取記憶體(聰(1〇1]1咖娜11^111()1^1^]^,$ 以儲存可供處理單it 11執行之程式模組以及資料。一般而言,程式模組包 含常序(routines)、程式扣嗯㈣、物件(object)、元件(c〇mp〇nent)等,用以 φ 執行電子學習順序編輯系統功能。本發明亦可以實施於分散式運算環境, • 其運算工作被一連結於通訊網路之遠端處理設備所執行。在分散式環境 中,基於知識本體之自動化呼吸波形判讀暨量測系統1〇之功能執行,也許 、 由本地以及多部遠端電腦系統共同完成。儲存裝置13包含硬碟裝置、軟碟 裝置、光碟裝置或隨身碟裝置,用以讀取硬碟、軟碟、光碟、隨身碟中儲 存之程式模組以及/或資料。 第2圖係表示依據本發明實施例之範例呼吸波形知識本體架構圖。呼 吸波形知識本體200可儲存於記憶體12(第丨圖)或儲存裝置13(第1圖)中, 疋一個由領域層210、主類別層23〇、次類別層250及概念層270,這四層 所組成的物件導向式知識本體架構。主類別層23〇包含壓力對時間波形圖、 容積對時間波形圖、流速對時間波形圖、壓力對容積波形圖以及流速對容 積波形圖等主類別。次類別層250包括呼吸時間可能性分佈、呼吸氣道壓 力可能性分佈、呼吸波形斜率可能性分佈等次類別。概念層27〇又由生理 參數層及症狀層273所組成。生理參數層271包含呼吸時間、呼吸氣 道壓力、波形斜率、肺順應性、肺部生理機轉、呼吸功、動脈血液氣體分 , 析等概念(c_ept),以及各概念間的關聯(associati〇ns)。症狀層273包含氣 %、慢性支氣管炎、支氣管擴張、塵肺症、肺纖維化等概念,以及各概念 間的關聯。Client's Docket No.: ACI94022 TT^ Docket No:0213-A40656-TW/Final/Jonah/20051117 8 1270363 Λ Microprocessor-based or programmable consumer electronics (miCr〇process〇r_baSed οι* consumer Electronics), network computers, mini computers, mainframes and similar devices. The processing unit 11 may comprise a single central processing unit (central_pr〇cessing_cpu) or a plurality of flat 5 line processing units associated with the parallel computing environment __processing. 3 Recalling body 12 contains read-only memory (1^〇11 to 11^111〇1^; 1^〇]^), flash memory (£^11〇]\4) and/or dynamic access memory体(聪(1〇1)1咖娜11^111()1^1^]^,$ to store program modules and data for processing single it11. In general, the program module contains the sequence (routines), program deduction (four), object (object), component (c〇mp〇nent), etc., for performing the e-learning sequence editing system function. The invention can also be implemented in a decentralized computing environment, • its computing work Executed by a remote processing device connected to the communication network. In a decentralized environment, the function of the automated breathing waveform interpretation and measurement system based on the knowledge ontology is performed, perhaps by local and multiple remote computer systems. The storage device 13 includes a hard disk device, a floppy disk device, a compact disk device or a flash drive device for reading a program module and/or data stored in a hard disk, a floppy disk, a compact disk, a flash drive, and the like. An exemplary breathing waveform knowledge ontology architecture diagram is provided according to an embodiment of the present invention. The respiratory waveform knowledge body 200 can Stored in the memory 12 (Fig. 1) or the storage device 13 (Fig. 1), an object consisting of the domain layer 210, the main class layer 23, the sub-category layer 250, and the concept layer 270. A guided knowledge ontology architecture. The main class layer 23 includes a pressure versus time waveform, a volume versus time waveform, a flow versus time waveform, a pressure vs. volume waveform, and a flow versus volume waveform. The subcategory layer 250 includes The respiratory time probability distribution, the respiratory airway pressure probability distribution, and the respiratory waveform slope probability distribution are sub-categories. The conceptual layer 27〇 is composed of a physiological parameter layer and a symptom layer 273. The physiological parameter layer 271 includes breathing time and respiratory airway pressure. , waveform slope, lung compliance, pulmonary physiology, respiratory work, arterial blood gas, analysis, etc. (c_ept), and the association between concepts (associati〇ns). Symptom layer 273 contains gas%, chronic bronchi Concepts such as inflammation, bronchiectasis, pneumoconiosis, pulmonary fibrosis, and the association between concepts.
Client 5s Docket N〇.:ACI94022 TT5s Docket No:〇213-A40656-TW/Final/J〇nah/20051117Client 5s Docket N〇.:ACI94022 TT5s Docket No:〇213-A40656-TW/Final/J〇nah/20051117
(I 7 1270363 第3圖係表示依據本發明實施例之範例之正常呼吸波形,其中包含起 始氣道壓力(airway pressure)、尖峰壓力加吐pressure)、高原壓力⑼拙如 pressure)、吐氣末正壓力 ^p〇skive end_expi加〇ry pressure)、吸氣時間 (inspiration time)及呼氣時間(eXpirati〇n time)等資訊。通常,每一個呼吸波形 包含六個特徵點A到F。 本發明之基於知識本體之自動化呼吸波形判讀暨量測方法分為兩階 段:判讀訓練;與實際量測。於判讀訓練階段中,提供相應於一個訓練呼 吸波形之特徵點(signiflcant points,如第3圖中之A到F)的壓力對時間資 訊、任意兩個特徵點間之呼吸氣道壓力可能性分佈㈧如仙办 distribution,or pressure membership也触⑽)與呼吸時間可能性分佈作咖 possibility distribution,or time membership function)資訊,據以計算出任意 兩個特徵點間之呼吸波形斜率可能性分佈(sl〇pe p〇ssibmty distributi〇n,沉 slope membership function)資訊,並建構出呼吸波形知識本體(亦可稱為訓練 知識本體)。於量測階段中,提供相應於一個實際呼吸波形之特徵點的壓力 對時間資訊,此實際呼吸波形可由一個呼吸量測儀器所偵測而得,代表一 個病患或被檢查者的呼吸波形。接著,依據判讀訓練階段所得到之任意兩 個特徵點間之呼吸氣道壓力、呼吸時間與呼吸波形斜率可能性分佈資訊, 計算並顯示實際呼吸波形中之任意一個特徵點之呼吸氣道壓力、呼吸時間 與任意兩個特徵點間之呼吸波形斜率的隸屬程度。當計算後之隸屬程度越 接近於”1”,代表實際呼吸波形中之特定特徵點之呼吸氣道壓力、呼吸日^間 與特定特徵點間之呼吸波形斜率,越趨近於正常的情況。反之,當計算後 之隸屬程度越偏離”1”,代表實際呼吸波形中之特定特徵點之呼吸氣道壓 力、呼吸時間與特定特徵點間之呼吸波形斜率,越偏離於正常的情況。兩 階段中之實施細節請參考以下的說明。 第4圖係表示依據本發明實施例之自動化呼吸波形判讀暨量測之判讀 訓練方法。於步驟S411,提供相應於一個訓練呼吸波形之特徵點的壓力對(I 7 1270363 Fig. 3 shows a normal respiratory waveform according to an example of an embodiment of the present invention, including initial airway pressure, peak pressure plus pressure, plateau pressure (9) such as pressure), and end-expiration Pressure ^p〇skive end_expi plus ry pressure), inspiration time and exhalation time (eXpirati〇n time) and other information. Typically, each breathing waveform contains six feature points A through F. The automatic ontology waveform interpretation and measurement method based on the knowledge ontology of the invention is divided into two stages: interpretation training; and actual measurement. During the interpretation training phase, provide pressure versus time information corresponding to a training breath waveform (signiflcant points, such as A to F in Fig. 3), and a respiratory airway pressure probability distribution between any two feature points (8) Such as the distribution, or pressure membership (10)) and the distribution of the possibility of breathing time, or time membership function information, according to the calculation of the probability distribution of the respiratory waveform between any two feature points (sl〇 Pe p〇ssibmty distributi〇n, sinking membership function), and constructing the respiratory waveform ontology (also known as training ontology). In the measurement phase, pressure versus time information corresponding to a feature point of an actual respiratory waveform is provided. The actual respiratory waveform can be detected by a respiratory measuring instrument to represent the respiratory waveform of a patient or subject. Then, based on the respiratory airway pressure, respiratory time and respiratory waveform slope probability distribution information between any two feature points obtained during the training phase, the respiratory airway pressure and breathing time of any one of the actual respiratory waveforms are calculated and displayed. The degree of membership of the slope of the respiratory waveform between any two feature points. When the calculated degree of membership is closer to "1", the respiratory wave pressure between the respiratory airway pressure and the specific feature point of the specific feature point in the actual respiratory waveform is closer to the normal situation. Conversely, the more the degree of membership after the calculation deviates from "1", the more the respiratory wave pressure, the breathing time, and the slope of the respiratory waveform between the specific feature points representing the specific feature points in the actual respiratory waveform deviate from the normal situation. Please refer to the following instructions for implementation details in the two phases. Fig. 4 is a diagram showing an automatic breathing waveform interpretation and measurement interpretation training method according to an embodiment of the present invention. In step S411, a pressure pair corresponding to a feature point of a training respiratory waveform is provided
Client’s Docket No·:ACI94022 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117 1270363Client’s Docket No·: ACI94022 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117 1270363
時間資訊。此資訊可以由-個檔案或資料庫中輸人,或可以透過一個使用 者介面(useri血face,m)輸入。第5圖係表示依據本發b月實施例之範例訓練 呼吸波形之特徵點的壓力對時間輸入介面則,其中包含六個子輸入區域 510至560,分別相應於如第3圖所示之特徵點A到F,每一個子輸入區域 可讓使用者輸人指定特徵_壓力與_等資訊。於步驟⑽,提供任音 兩個特徵點間之呼吸氣道壓力可紐分佈魏。任意兩個特徵關之啊 氣道壓力可能性分佈資訊包含模糊數的起始支持师响卿⑽肩、起 始核心點(begin㈣,BC)、結束核心點_賺,EC)與結束支持點_ 哪P〇rt,ES)。此資訊可由步驟則所提供的資訊計算而得,由一個播案或 資料庫中輸人’或可以透過-個使用者介面輸人。第6a圖係表示依據本發 明實施例之範鑛顏A卿_ 3 _啊氣道 介細,包含四個輸入攔細、613、615與617,分概讓 入模糊數”呼魏道壓力”的起始域點、起触_、結束如輯处束 支持點。於步驟S433,提供任意兩個特徵點間之呼吸時間可能性分佈資L 任意兩個特徵點狀呼吸_可紐分佈f訊包含翻數的起始支持點、 起始核心點、結束核心點與結束支持點。此資訊可由步驟s4u所提供的資 訊計算而得,由-健案或資料庫中輸入,或可以透過—個使用料面輸 入。第6b圖係表示依據本發明實施例之範例特徵點a到b間(第3靶的: 吸時間可能性分佈輸入介面630,包含四個輸入搁位631、633 Μ% :纪7, 分別用來讓使用者輸入模糊數,,呼吸時間,,的起始支持點、起始核心點、社 束核心點與結束支持點。 人"〜 於步驟S451,依據呼吸氣道壓力與時間可能性分佈資訊,吁曾出任立 兩個特徵闕之呼吸波形斜率可能性分佈資訊。任細個特徵 波形斜率可紐分鍾訊包含模驗的起始支持點、起始核心點、社束核 心點與結束支持點。此步驟可使用模糊波形斜率建構演算法Time information. This information can be entered from a file or database, or it can be entered via a user interface (useri blood face, m). Figure 5 is a diagram showing the pressure versus time input interface of the feature points of the training breathing waveform according to the example of the present invention, including six sub-input areas 510 to 560, corresponding to the feature points as shown in Fig. 3, respectively. A to F, each sub-input area allows the user to enter information such as characteristics _pressure and _. In step (10), the respiratory airway pressure between the two feature points is provided. Any two features related to the airway pressure probability distribution information including the starting number of the fuzzy number of supporters (10) shoulder, starting core point (begin (four), BC), ending core point _ earning, EC) and ending support point _ P〇rt, ES). This information can be calculated from the information provided in the step, from a broadcast or database to the input or can be entered through a user interface. Fig. 6a is a diagram showing the embodiment of the invention, according to the embodiment of the present invention, the squadron A _ 3 _ ah airway fine, including four input barriers, 613, 615 and 617, the subdivision of the fuzzy number "Hu Weidao pressure" The starting domain point, the start _, and the end point are the bundle support points. In step S433, the respiratory time probability distribution between any two feature points is provided. L. Any two characteristic point-like breaths _ can be distributed, including the starting support point, the starting core point, and the ending core point of the flip number. End the support point. This information can be calculated from the information provided in step s4u, entered from the health record or database, or can be entered through a use surface. Figure 6b shows an exemplary feature point a to b according to an embodiment of the present invention (the third target: suction time likelihood distribution input interface 630, comprising four input shelves 631, 633 Μ%: 纪7, respectively To let the user input the fuzzy number, the breathing time, the starting support point, the starting core point, the social beam core point and the ending support point. The person "~ in step S451, according to the respiratory airway pressure and time probability distribution Information, Yu has served as the two characteristics of the respiratory waveform slope probability distribution information. Any fine characteristic waveform slope can be used to include the initial support point of the model, the starting core point, the social core point and the end support Point. This step can use the fuzzy waveform slope construction algorithm.
Waveform Slope Construction Algorithm)來計算出任意兩個特徵點間之]^Waveform Slope Construction Algorithm) to calculate between any two feature points ^
Client’s Docket No···ACI94022 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117 9 1270363Client’s Docket No···ACI94022 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117 9 1270363
波形斜率可能性分佈資訊,模糊波形斜率建構演算法如下所示。 輪入· Pbs,Pbc,Pec,Pes,TBS,TBC,TEC,TES 輪出:〜,,〜 步麻1:計算出呼吸波形斜率的四個可能起始核心點與結束核心點:Waveform slope probability distribution information, fuzzy waveform slope construction algorithm is shown below. Rounds · Pbs, Pbc, Pec, Pes, TBS, TBC, TEC, TES Round: ~,, ~ Step 1 : Calculate the four possible starting core points and ending core points of the respiratory waveform slope:
PbcITbc,pbcItec,PecItbc,PecItec 步麻2:計算出呼吸波形斜率的四個可能起始支持點與結束支持點:PbcITbc, pbcItec, PecItbc, Pecitec Step 2: Calculate the four possible starting and ending support points for the slope of the respiratory waveform:
PbS / ^BS 5 ^BS / TES ? PES / TBS , PES / TES 步麻3:取得呼吸波形斜率的起始支持點、起始核心點、結束核心點 與結束支持點·· sBS 二 Mm、PBS/TBS, pbsItes, pjTbs, pjTe“ sBC =MIN(p,c/r5C5 pbc!tec, pec/tbc, pec/Tec) sEC =MAX{pbc/Tbc, pbc!tec, pec/Tbc, pec/Tec) sES -MAX(p^/r^5 pbsItes, pes/tbs^ pes/tes) 步驟4:結束 演异法巾u減的輸人參數,表賴驗,,啊氣賴力,,之起 支持點;P5C為程式的輸入參數,表示模糊數,,呼吸氣道遷力,,之起始核 點;〜為程式的輸入參數,表示模糊數’,令及氣道壓力,,之結束核心點;7 為程式的輸人錄,表稍_,’呼魏碰力”讀束支賊為程 的輸4數,表示模糊數,,呼吸時間,,之起始支持點、為程式的輸入參數 糊數呼辦間之触核為程式的輪人參數,表示模糊藝 鱼,吸日扣間’’之結束核心點、為程式的輸出參數,表示模糊數,,呼吸波辦 “ σ支持點’〜為転式的輸出參數,表示模細:,,呼吸波形斜率,,之夫 始核心點、為程式的輸出參數,表示模糊數,,呼吸波形斜率,,之結束如 ^ 6^1 雜摘輸岭數,絲翻數,,呼赠形解,,之絲支持點< =c圖絲__㈣魏狀__ A _ 形斜率可紐分佈顯示介面65G,包含四個顯示欄位65i、653、65)5的^PbS / ^BS 5 ^BS / TES ? PES / TBS , PES / TES Step 3 : Get the starting support point , starting core point , ending core point and ending support point of the respiratory waveform slope · · sBS Mm, PBS /TBS, pbsItes, pjTbs, pjTe" sBC =MIN(p,c/r5C5 pbc!tec, pec/tbc, pec/Tec) sEC =MAX{pbc/Tbc, pbc!tec, pec/Tbc, pec/Tec) sES -MAX(p^/r^5 pbsItes, pes/tbs^ pes/tes) Step 4: End the different input parameters of the different method, the table depends on the test, the ah, the gas, the support point P5C is the input parameter of the program, indicating the fuzzy number, the respiratory airway migration force, and the starting nuclear point; ~ is the input parameter of the program, indicating the fuzzy number ', the air pressure, and the ending core point; 7 is The program's input record, the table slightly _, 'Hu Wei touch force' read the bundle thief for the process of losing 4 numbers, indicating the fuzzy number, breathing time, the starting support point, the input parameter of the program The touch between the office is the program's wheelman parameter, which means that the fuzzy art fish, the end point of the 'day button' is the end of the program, the output parameter of the program, indicating the fuzzy number, and the respiratory wave is “σ support point”~ formula The output parameters indicate the modulus:,, the slope of the respiratory waveform, the starting core point of the husband, the output parameter of the program, the fuzzy number, the slope of the respiratory waveform, and the end of the cycle, such as ^ 6^1丝翻数,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 5 of ^
Client’s Docket No·:ACI94022 wke⑽細.Α4·τ卿讀_2_i7 10Client’s Docket No·: ACI94022 wke(10)细.Α4·τ卿读_2_i7 10
Ci 1270363 分別用來顯示計算後之模糊數”呼吸波形斜率,,的起始支持點 點、結束核心點與結束支持點。 、…起始核心 於步驟S47;l,建構相應於呼吸波形知識本體2〇〇(第2圖)的呼吸波形 知識庫(kn〇wledge base)。知識庫中包含提供之任意兩個特徵點間之啤吸氣 與^日可能性分佈資訊,以及利_波形斜率建構演算法所 外异出任忍兩個特徵點間之呼吸波形斜率可能性分佈資訊。 旦、第7 _表示依據本發明實施例之自動化呼吸波形判讀暨量測之實際 ==法。^步驟S7U,提供相應於一個實際呼吸波形之特徵點的壓“ 夺間貝戒。舉例來說,提供之呼吸波形的特徵點(Z7,...,jr„),尤=戶7 ,其中Ρι,·.·,ρ”中之—者代表—個特定特徵點之呼吸氣道(=”而’ U中之—者代表—㈣定特徵點之,瞒咖。此魏可以由一個標 案或資料庫中輸人,或可以透過—個使用者介面輸人。第如_ : 本發明實施例之·實際呼讀形之特徵闕壓力料嶋人介面⑽(盆 ΙΓΓ:子輸入區域811、813、815與817,每-個子輸入區域可咖 者輸入扣疋特徵點的壓力與時間等資訊。 參考第7 ®,於步驟S731,依據相應之任意兩個特徵 ,之I能性分佈資訊(第4圖中之步驟⑽所示),取_ = 中1意-個特徵點之呼吸氣道勤之隸屬程度。於步驟咖,依手據= 之任忍兩個特徵點間之呼吸時間之可能性分佈資訊(如第4圖 所示),取碰力對時間資訊中之任意一個特徵點之呼 程 於步驟S751,計算勤對時間資訊中之任意兩個特徵點間之呼吸 (sl_,任意兩個特徵點間之呼吸波形斜村使用如下的 ^ ^步驟S753 ’依據域之__概闕之啊 資訊(如第4财之步驟S435所示),取懸 =之了犯丨生刀佈 徵點間之呼吸波科率之·程度 、=财之任意兩個特 反於步驟S77;I,透過輸出單元14(第ΪCi 1270363 is used to display the calculated fuzzy number "respiratory waveform slope", the initial support point, the end core point and the end support point, respectively. The starting core is in step S47; l, constructing the corresponding body corresponding to the respiratory waveform 2〇〇 (Fig. 2) Respiratory waveform knowledge base (kn〇wledge base). The knowledge base contains information on the distribution of beer inhalation and probability of any two feature points provided, and the construction of the slope of the profit_waveform The algorithm is different from the residual waveform of the respiratory waveform between the two feature points. The seventh _ represents the actual == method of the automatic respiratory waveform interpretation and measurement according to the embodiment of the present invention. ^ Step S7U, A pressure "between ring" corresponding to a characteristic point of an actual respiratory waveform is provided. For example, the characteristic points (Z7,...,jr„) of the respiratory waveform provided, especially = household 7, where Ρι,·.·, ρ" represent the respiratory airway of a particular feature point ( = "And 'U' - the representative - (four) fixed feature points, 瞒 。. This Wei can be entered by a standard or database, or can be input through a user interface. For example _ : Ben In the embodiment of the invention, the characteristics of the actual call-reading form are as follows: the pressure input device (10) (the basin: the sub-input areas 811, 813, 815 and 817, the pressure and time of the 可 疋 疋 每 每 每 每 每 每 每 每 每 每 疋 疋 疋 输入For information, refer to Section 7®, in step S731, according to any two features corresponding to the I energy distribution information (shown in step (10) in Figure 4), take _ = 1 meaning - a characteristic point of breathing The degree of subordination of airway diligence. In the step coffee, according to the data of the hand, the information on the possibility of breathing time between the two feature points (as shown in Figure 4), taking any of the features of the force versus time information The call of the point is in step S751, and the breathing between any two feature points in the time information is calculated (sl_, any The respiratory waveform between the two feature points is used as follows: ^ ^Step S753 'According to the domain __ Overview of the information (as shown in step 4 of the fourth wealth), take the suspension = the knives The degree of the respiratory wave rate between the points, any two of the money is opposite to the step S77; I, through the output unit 14 (the third
Client’s Docket N〇.:ACI94022 TT^s Docket N〇:0213-A40656-TW/Final/J〇nah/20051117 1270363 '圖)顯示各隸屬程度之計算結果於—個顯示器上。第Sb圖係表示依據本發 ' Θ實蝴之1&例拉結果顯讀面83G,其巾包含特徵點A(如第2圖所示) 之呼吸氣道壓力之隸>|赌⑶、點B(如帛2 _椒啊氣道壓力 ^隸屬程度833、特徵點a之呼吸時間之隸屬程度835、特徵點B之啤吸 欠間之隸屬粒度837與特徵點a與B間之呼吸波形斜率之隸屬程度等 -貝訊。具體來說,當輸入呼吸波形特徵點A=(7 5, 〇 25)及特徵點b=(22 5, 〇 3) 時,則特徵點A隸胁這讎力可能性分佈及時間的可能性分佈的程度是 1、特徵點B隸屬於這健力可紐分佈及時關可齡㈣的程度也是 # 1(參考第如與你圖)’而特徵點A到特徵點B所形成的斜率,隸屬於這個 斜率可能性分佈的程度也是K參考第6c圖)。 $ 9 ® 實施例之基於知識本體之自動化呼吸波形判 讀暨量測之電腦可讀取儲存媒體示意圖。此儲存媒體9〇,用以儲存一電腦 程式92G,狀實觀上所叙知識本社自動化呼磁義讀暨量測 方法(包含繼娜與實㈣财法)。本發明之方法與纽,如校型態或 其部份,可以以程式碼的型態包含於實體媒體,如軟碟、光碟片、硬碟、 或是任何其他機器可讀取(如電腦可讀取)儲存媒體,其中,當程式碼被機 • 15,如電腦載入且執行時’此機器變成用以參與本發明之裝置。本發明之 方法與裝置也可以以程式碼型態透過一些傳送媒體,如電線或魏、光纖、 或是任何傳輸型態進行傳送,其中,當程式碼被機器,如電職收、載入 且執行時,此機器變成用以參與本發明之裝置。當在一般用途處理單元 (general-pu—ep職sing_實作時’程式碼結合處理器提供一操作類似 於應用特定邏輯電路之獨特裝置。 雖然本發明已讀佳實關跡如上,然其並_錄定本發明,任 何熟悉此項技藝者,在不脫離本發明之精神和範圍内,當可做些許更動盘 ΙίιϊϋΙΓ】之賴細#紐社㈣專鄕麟狀者為準。”Client's Docket N〇.: ACI94022 TT^s Docket N〇:0213-A40656-TW/Final/J〇nah/20051117 1270363 'Figure' shows the calculation results of each degree of membership on a display. The Sb diagram shows the appearance of the reading surface 83G according to the present invention. The towel contains the characteristic point A (as shown in Fig. 2) of the respiratory airway pressure >|gambling (3), point B (such as 帛2 _ pepper ah airway pressure ^ degree of membership 833, characteristic point a of the respiratory time of the degree of membership 835, feature point B of the beer owing between the sub-granularity 837 and the characteristic point a and B between the respiratory waveform slope Degree of membership, etc. - Beyond. Specifically, when the input respiratory waveform feature points A = (7 5, 〇 25) and the feature points b = (22 5, 〇 3), then the feature point A is threatened by this force. The degree of probability distribution of sexual distribution and time is 1. The degree of characteristic point B belongs to this Jianli can be distributed in time and the age of (4) is also #1 (refer to the picture as you figure) and the characteristic point A to the characteristic point B The slope formed, to the extent that this slope is likely to be distributed, is also referred to in Figure 6c. $ 9 ® Example of a computer-readable storage medium based on knowledge ontology for automated respiratory waveform interpretation and measurement. This storage medium is 9〇, which is used to store a computer program 92G. It is a description of the knowledge of the Society's automated magnetic reading and measurement methods (including Ji Na and Shi (4) Finance). The method and the button of the present invention, such as a school type or a part thereof, may be included in a physical medium such as a floppy disk, a CD, a hard disk, or any other machine (such as a computer). The storage medium is read, wherein when the code is accepted by the machine, 15, if the computer is loaded and executed, the machine becomes a device for participating in the present invention. The method and apparatus of the present invention may also be transmitted in a code format through some transmission medium such as a wire or a fiber, an optical fiber, or any transmission type, wherein when the code is loaded by the machine, such as electricity, When executed, the machine becomes a device for participating in the present invention. When the general-purpose processing unit (general-pu-ep-sing_implementation), the code-integrated processor provides a unique device that operates similarly to the application-specific logic circuit. Although the present invention has read the good-looking trace as above, And </ RTI> </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt;
Clienfs Docket No. :ACI94022 TT s Docket No:0213-A40656-TW/Final/Jonah/20051117Clienfs Docket No. :ACI94022 TT s Docket No:0213-A40656-TW/Final/Jonah/20051117
12 1270363 •、 第1圖係表示依據本發明實施例之基於知識本體之自動化呼吸波形判 . 讀暨量測之硬體架構圖; 第2圖係表示依據本發明實施例之範例呼吸波形知識本體架構圖; 第3圖係表示依據本發明實施例之範例之正常呼吸波形; 第4圖係表示依據本發明實施例之自動化呼吸波形判讀暨量測之判讀 訓練方法; 第5圖係表示依據本發明實施例之範例訓練呼吸波形之特徵點的壓力 對時間輸入介面; # 第6a圖係表示依據本發明實施例之範例特徵點A到B間的呼吸氣道 壓力可能性分佈輸入介面; 第6b圖係表示依據本發明實施例之範例特徵點a到B間的呼吸時間 可能性分佈輸入介面; 弟6c圖係表示依據本發明實施例之範例特徵點a到B間的啤吸波形 斜率可能性分佈顯示介面; 第7圖係表示依據本發明實施例之自動化呼吸波形判讀暨量測之實際 量測方法; 第8a圖係表示依據本發明實施例之範例實際呼吸波形之特徵點的壓 籲力對時間輸入介面; 第8b圖係表示依據本發明實施例之範例計算結果顯示介面; 第9圖係表示依據本發明實施例之基於知識本體之自動化呼吸波形判 讀暨量測之電腦可讀取儲存媒體示意圖。 【主要元件符號說明】 10〜自動化呼吸波形判讀暨量測系統;11〜處理單元;12〜記憶體;13〜 儲存裝置;14〜輸出裝置;15〜輸入裝置;16〜通訊裝置;17〜匯流排;200〜 呼吸波形知識本體;210〜領域層;230〜主類別層;250〜次類別層;270〜相无 念層;271〜生理參數層;273〜症狀層;A、B、C、D、E、F〜特徵點;S411、12 1270363 • Fig. 1 is a diagram showing the hardware structure of the automatic respiratory waveform judgment and reading based on the knowledge ontology according to an embodiment of the present invention; and Fig. 2 is a diagram showing an example respiratory waveform knowledge ontology according to an embodiment of the present invention. FIG. 3 is a diagram showing a normal respiratory waveform according to an example of an embodiment of the present invention; FIG. 4 is a diagram showing an automatic breathing waveform interpretation and measurement interpretation training method according to an embodiment of the present invention; Example of the embodiment of the invention trains the pressure versus time input interface of the feature points of the respiratory waveform; #图 6a shows the respiratory airway pressure probability distribution input interface between the example feature points A to B according to an embodiment of the invention; The breathing time likelihood distribution input interface between the example feature points a to B according to the embodiment of the present invention is shown; the brother 6c diagram shows the probability distribution of the beer suction waveform slope between the example feature points a to B according to the embodiment of the present invention. Display interface; Figure 7 shows an actual measurement method for automatic respiratory waveform interpretation and measurement according to an embodiment of the present invention; Figure 8a shows The pressing force versus time input interface of the characteristic point of the actual respiratory waveform according to the embodiment of the present invention; the 8b is an example calculation result display interface according to an embodiment of the present invention; and the ninth figure is the embodiment according to the embodiment of the present invention. A schematic diagram of a computer readable storage medium based on knowledge ontology for automated respiratory waveform interpretation and measurement. [Main component symbol description] 10~Automatic respiratory waveform interpretation and measurement system; 11~ processing unit; 12~memory; 13~ storage device; 14~output device; 15~ input device; 16~ communication device; Row; 200~ respiratory waveform knowledge ontology; 210~ domain layer; 230~ main category layer; 250~ sub-category layer; 270~ phase non-layer; 271~ physiological parameter layer; 273~ symptom layer; A, B, C, D, E, F ~ feature points; S411,
Client’s Docket N〇.:ACI94022 13 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117 1270363 … S43卜S433、S451、S47l〜方法步驟;500〜特徵點的壓力對時間輪入介面; 510、520、530、540、550、560〜子輸入區域;610〜呼吸氣道壓力可能性分 佈輸入介面;611、613、615、617〜子輸入欄位;630〜呼吸時間可能性分佈 輸入介面;63卜633、635、637〜子輸入欄位;650〜呼吸波形斜率可能性分 佈顯示介面;651、653、655、657〜顯示欄位;S711、S731、S733、S751、 S753、S771〜方法步驟;810〜特徵點的壓力對時間輸入介面;81丨、813、815、 817〜子輸入區域;830〜計算結果顯示介面;831〜特徵點A之呼吸氣道壓力 之隸屬程度;833〜特徵點B之呼吸氣道壓力之隸屬程度;835〜特徵點a之 • 呼吸時間之隸屬程度;837〜特徵點B之呼吸時間之隸屬程度;839嗜徵點 • A與B間之呼吸波形斜率之隸屬程度;9〇〜健存媒體;92〇〜基於知識本體之 自動化呼吸波形判讀暨量測電腦程式。Client's Docket N〇.: ACI94022 13 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117 1270363 ... S43 Bu S433, S451, S47l~ method steps; 500~ feature point pressure versus time rounding interface; 510, 520, 530, 540, 550, 560~ sub-input area; 610~ respiratory airway pressure possibility distribution input interface; 611, 613, 615, 617~ sub-input field; 630~ breathing time possibility distribution input interface; 63 633, 635, 637~ sub-input field; 650~ respiratory waveform slope probability distribution display interface; 651, 653, 655, 657~ display field; S711, S731, S733, S751, S753, S771~ method step; ~ characteristic point pressure versus time input interface; 81丨, 813, 815, 817~ sub-input area; 830~ calculation result display interface; 831~ feature point A's degree of respiratory airway pressure; 833~ feature point B breath Degree of membership of airway pressure; 835~feature point a • degree of subordination of breathing time; degree of subordination of breathing time of 837~feature point B; 839 lust point • degree of membership of respiratory waveform slope between A and B; ~健存Media; 92〇~ Based on the knowledge ontology Automatic breathing waveform interpretation and measurement computer program.
Client’s Docket N〇_:ACI94022 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117Client’s Docket N〇_: ACI94022 TT5s Docket No:0213-A40656-TW/Final/Jonah/20051117
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW94145529A TWI270363B (en) | 2005-12-21 | 2005-12-21 | Systems and methods for automated ventilator waveform recognition and measure based on ontologies |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW94145529A TWI270363B (en) | 2005-12-21 | 2005-12-21 | Systems and methods for automated ventilator waveform recognition and measure based on ontologies |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI270363B true TWI270363B (en) | 2007-01-11 |
TW200724093A TW200724093A (en) | 2007-07-01 |
Family
ID=38430104
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW94145529A TWI270363B (en) | 2005-12-21 | 2005-12-21 | Systems and methods for automated ventilator waveform recognition and measure based on ontologies |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI270363B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI447680B (en) * | 2012-07-05 | 2014-08-01 | Univ Nat Chiao Tung | Method and system on detecting abdominals for singing |
TWI476723B (en) * | 2013-07-25 | 2015-03-11 | Univ Nat Chiao Tung | Personalized system and method of abdominal breathing training evaluation based on abdominal muscles cluster function |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW201544074A (en) * | 2014-05-22 | 2015-12-01 | Apex Medical Corp | Breathing waveform recognition method and system thereof |
-
2005
- 2005-12-21 TW TW94145529A patent/TWI270363B/en active
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI447680B (en) * | 2012-07-05 | 2014-08-01 | Univ Nat Chiao Tung | Method and system on detecting abdominals for singing |
TWI476723B (en) * | 2013-07-25 | 2015-03-11 | Univ Nat Chiao Tung | Personalized system and method of abdominal breathing training evaluation based on abdominal muscles cluster function |
Also Published As
Publication number | Publication date |
---|---|
TW200724093A (en) | 2007-07-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Al Osman et al. | Ubiquitous biofeedback serious game for stress management | |
McGregor | Big data in neonatal intensive care | |
Daniels et al. | Accurate assessment of adherence: self-report and clinician report vs electronic monitoring of nebulizers | |
US8696592B2 (en) | Breath biofeedback system and method | |
Burkhardt et al. | The diagnosis of chronic obstructive pulmonary disease | |
Mottram | Ruppel's Manual of Pulmonary Function Testing10: Ruppel's Manual of Pulmonary Function Testing | |
EP3365057A1 (en) | System and method for monitoring and determining a medical condition of a user | |
CN111000541A (en) | Method, system and device for making and implementing personalized deep breathing training prescription | |
Magalang et al. | Agreement in the scoring of respiratory events among international sleep centers for home sleep testing | |
WO2019061941A1 (en) | Traditional chinese medicine inquiry apparatus and traditional chinese medicine inquiry data processing device | |
CN114974613A (en) | Disease management method and device, computer storage medium and electronic equipment | |
Chatburn et al. | 2019 year in review: patient-ventilator synchrony | |
Sanchez-Perez et al. | A wearable multimodal sensing system for tracking changes in pulmonary fluid status, lung sounds, and respiratory markers | |
CN111403041A (en) | Diagnosis process simulation system based on virtual standard patient | |
Hawkes et al. | Auscultate, palpate and tap: time to re‐evaluate | |
Wu et al. | Development of quality assurance and quality control guidelines for respiratory oscillometry in clinic studies | |
TWI270363B (en) | Systems and methods for automated ventilator waveform recognition and measure based on ontologies | |
Mishlanov et al. | Scope and new horizons for implementation of m-Health/e-Health services in pulmonology in 2019 | |
CN111402982A (en) | Diagnosis evaluation system based on virtual standard patient | |
EP4072414B1 (en) | System and method for metabolic measurements | |
Patel et al. | Refill rates of accessories for positive airway pressure therapy as a surrogate measure of long-term adherence | |
Jayasekera et al. | Feasibility of using the Hexoskin smart garment for natural environment observation of respiration topography | |
TW201227382A (en) | Health activity plan establishment system, establishment method, and its computer program product | |
Chen et al. | The intention to use the cloud sphygmomanometer-demonstrated by Taiwan medical center | |
TWI505812B (en) | System and method for displaying analysis of breath |