WO2021090659A1 - Dispositif, procédé et programme permettant de prédire l'effet thérapeutique d'une thérapie de resynchronisation cardiaque - Google Patents

Dispositif, procédé et programme permettant de prédire l'effet thérapeutique d'une thérapie de resynchronisation cardiaque Download PDF

Info

Publication number
WO2021090659A1
WO2021090659A1 PCT/JP2020/038910 JP2020038910W WO2021090659A1 WO 2021090659 A1 WO2021090659 A1 WO 2021090659A1 JP 2020038910 W JP2020038910 W JP 2020038910W WO 2021090659 A1 WO2021090659 A1 WO 2021090659A1
Authority
WO
WIPO (PCT)
Prior art keywords
resynchronization therapy
cardiac resynchronization
responder
patient
prediction
Prior art date
Application number
PCT/JP2020/038910
Other languages
English (en)
Japanese (ja)
Inventor
悠介 関
悠乃 北川
周平 松下
Original Assignee
テルモ株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by テルモ株式会社 filed Critical テルモ株式会社
Publication of WO2021090659A1 publication Critical patent/WO2021090659A1/fr

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N1/00—Electrotherapy; Circuits therefor
    • A61N1/18—Applying electric currents by contact electrodes
    • A61N1/32—Applying electric currents by contact electrodes alternating or intermittent currents
    • A61N1/36—Applying electric currents by contact electrodes alternating or intermittent currents for stimulation
    • A61N1/362—Heart stimulators

Definitions

  • the present invention relates to a cardiac resynchronization therapy therapeutic effect predictor, a method and a program for predicting the therapeutic effect of cardiac resynchronization therapy.
  • heart failure is defined as "some kind of cardiac dysfunction, that is, dyspnea, malaise, and edema appear as a result of an abnormality in the heart that disrupts the compensatory function of the cardiac pump function, and exercise tolerance is associated therewith. It is defined as “clinical syndrome with diminished ability”. In other words, heart failure does not refer to the name of the disease, but to "the state of the body that has occurred as a result of insufficient heart function.” There are various causes of heart failure such as heart attack, hypertension, valvular disease, myocarditis, congenital cardiovascular system, and arrhythmia. Therefore, an appropriate treatment method is selected from drug treatment, surgery, and device treatment according to the cause.
  • Intraventricular conduction disorders are usually perceived as an extension of the QRS complex of the electrocardiogram. About one-third of cases of chronic heart failure show QRS complexes of 120 ms or longer, most of which are reported to be left bundle branch block. Cardiac resynchronization therapy (CRT) is effective for heart failure caused by such ventricular synchronization failure.
  • CRT Cardiac resynchronization therapy
  • CRT is an effective treatment for severe heart failure.
  • CRT is a device therapy to make ventricular contraction more systematic and effective.
  • CRT also called biventricular pacing, maximizes the amount of blood pumped by the heart by simultaneously contracting both ventricles by electrical stimulation of the left and right ventricles.
  • CRT implantation surgery is performed under local anesthesia, similar to normal pacemaker implantation surgery.
  • the CRT device is implanted subcutaneously in the chest below the clavicle.
  • Three leads are inserted into the ventricle along the subclavian vein under fluoroscopy, and the electrodes at the tips of the two leads are contact-fixed to the inner walls of the right atrium and right ventricle.
  • the third lead wire is inserted into the confluent vein of the coronary vein running on the surface of the heart called the coronary sinus that joins the right atrium, and the lead wire crawls into this coronary vein in the left ventricle.
  • a lead wire running on the surface is inserted and placed intravenously.
  • Non-Patent Document 1 the recommendation and evidence level of CRT are "1.
  • Optimal drug treatment 2. Left Ventricular” for NYHA (New York Heart Association) functional classification III / IV degree.
  • the NYHA functional classification II degree is "1.
  • Optimal drug treatment 2. LVEF ⁇ 35%, 3.
  • Non-Patent Document 1 An index for evaluating a patient who can be a non-responder, such as the size of the QRS width, has been conventionally proposed.
  • the QRS complex is the waveform of the electrocardiogram generated by the contraction of the ventricles, and the QRS width is the length of the appearance time of the QRS complex.
  • the normal QRS width is less than 120 ms.
  • conduction failure such as left bundle branch block occurs, the QRS width tends to increase due to conduction delay.
  • a wide QRS complex indicates that the pumping function of the heart may be impaired, as delayed conduction in the ventricles causes a time lag in contraction of the ventricles.
  • Non-Patent Document 2 uses data from a multicenter clinical study (MADIT-CRT) on the application of an implantable cardioverter defibrillator (CRT-D) with biventricular pacing function, and 677 patients who received CRT-D.
  • ICD implantable cardioverter defibrillator
  • Non-Patent Document 1 CRT is recommended for patients with a QRS width of 120 ms or more on the electrocardiogram.
  • QRS complex of less than 120 ms, and about half of them have ventricular dyssynchrony on echocardiography, so CRT may also be effective in these patients. Therefore, it is required to predict the therapeutic effect of CRT regardless of indicators such as QRS complex and LVEF.
  • Non-Patent Document 2 targets patients with mild heart failure, and non-responders of patients with severe heart failure cannot be classified.
  • An object of the present invention is to provide a device, a method and a program capable of predicting a non-responder of CRT regardless of the degree of symptoms of heart failure.
  • the cardiac resynchronization therapy therapeutic effect predictor of the present invention is medical information about a patient who has undergone cardiac resynchronization therapy including at least one of a responder and a non-responder of cardiac resynchronization therapy.
  • a teacher data input unit that accepts input of teacher data
  • a machine learning unit that machine-learns the prediction criteria of the non-responder based on the teacher data that the teacher data input unit has received input
  • a predictor unit for predicting whether or not the patient who has not undergone cardiac resynchronization therapy is a non-responder of cardiac resynchronization therapy is provided.
  • the computer is a teacher data which is medical information about a patient who has undergone cardiac resynchronization therapy including at least one of a responder and a non-responder of cardiac resynchronization therapy.
  • the patient who has not received the cardiac resynchronization therapy is the non-responder of the cardiac resynchronization therapy. And the step of predicting whether or not it is.
  • the present invention also includes a program for executing the above method.
  • the non-responder of CRT can be predicted with high accuracy by the present invention, only patients who can expect a therapeutic effect can receive CRT. This allows the non-responder to avoid the physical, mental, or financial burden associated with the resulting ineffective CRT procedure.
  • non-responders can have the opportunity to study and select a more effective treatment for heart failure in advance. Therefore, the present invention is expected to contribute to improving the QOL of non-responders.
  • the present invention contributes to the reduction of non-responders of CRT, it is possible to achieve a medical economic effect of reducing the medical expenses spent on unnecessary CRT treatment.
  • FIG. 1 is a block diagram of a cardiac resynchronization therapy therapeutic effect prediction device 1 according to a preferred embodiment of the present invention.
  • the cardiac resynchronization therapy treatment effect prediction device 1 includes a teacher data input unit 101, a machine learning unit 102, a medical information input unit 103, a prediction unit 104, and a prediction result output unit 105.
  • the cardiac resynchronization therapy treatment effect prediction device 1 can be configured by a computer having a well-known hardware configuration such as a CPU (Central Processing Unit), memory, storage, input / output interface, communication interface, and data bus. ..
  • CPU Central Processing Unit
  • the teacher data input unit 101 accepts input of medical information (teacher data) of a patient who is known to be a non-responder and / or a patient who is known to be a responder.
  • the teacher data input unit 101 includes MT (Magnetic Tape: magnetic tape), CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Why Disk Read Online Memory), USB (Universal Memory), USB (Universal), USB (Universal), and USB (Universal). It can be configured with known hardware such as a reading device for a recording medium such as Digital Card (Secure Digital Card). Alternatively, the teacher data input unit 101 may accept input of teacher data via a secure communication device.
  • the teacher data is medical information of a patient who is confirmed to be a non-responder and / or a patient who is confirmed to be a responder as a result of CRT treatment, that is, patient attribute information (age, age, Gender, weight, height, blood pressure, etc.), medical history, medication history, cardiac morphology (CT, echocardiography, etc.), electrical cardiac function (electrocardiogram, etc.), mechanical cardiac function (echocardiography, etc.), myocardial viability (myocardiography, etc.) It is at least one of the elements of scintigraphy, contrast MRI, etc.), lead placement site (X-ray fluoroscopic image, etc.), and pacing electrode placement position. It is desirable to have as many elements of teacher data and patients as sources of teacher data as possible. Therefore, the element of medical information in the teacher data does not have to be limited to the above.
  • the teacher data may include elements such as LVEF, QRS width, and NYHF.
  • the teacher data includes CT showing anatomical information of the myocardium, guided electrocardiogram showing mechanical synchronization failure, myocardial scintigraphy and contrast-enhanced MRI showing the viability of the myocardium, and X-ray fluoroscopic image showing the lead / electrode placement site. It is desirable to include it.
  • the teacher data may include elements of medical information of the same patient measured at different dates and times.
  • It may include body weight, cardiac morphology, electrical cardiac function, and the like.
  • the teacher data may include medical information of a patient who has undergone CRT but cannot obtain a therapeutic effect. For such patients, changing the electrode placement of the CRT may result in a responder.
  • a patient may be referred to as a "provisional non-responder”.
  • the teacher data may include medical information of the provisional non-responder. The definition of "temporary non-responder" will be described later.
  • the machine learning unit 102 learns a standard for predicting whether or not a patient is a non-responder based on medical information (teacher data) for learning input to the teacher data input unit 101. That is, the machine learning unit 102 machine-learns the non-responder prediction determination model from the input teacher data.
  • the prediction judgment model a known one such as logistic regression or a support vector machine is used.
  • Image information such as heart morphology, myocardial viability, and lead placement site may be processed by GPU (Graphics Processing Unit) instead of CPU or in combination with CPU.
  • PLD Programmable Logic Device
  • FPGA Field-Programmable Gate Array
  • ASIC Application Specific Integrated Circuit
  • the machine learning unit 102 may machine-learn the prediction determination model of the temporary non-responder based on the teacher data of the temporary non-responder. It is even better that the machine learning unit 102 also machine-learns a model that predicts and determines the change position of the electrode in which the temporary non-responder becomes the responder.
  • the machine learning unit 102 predicts the mechanical and electrical cardiac functions of the patient at a predetermined period after the treatment based on the teacher data including the elements of the medical information of the same patient measured at different dates and times. It is even better to machine-learn the model to be judged.
  • the teacher data may be medical information about non-responders, medical information about responders, or both. This is because the predictor judgment model of the responder is machine-learned from the medical information about the responder, and predicting that the patient is not a responder is the same as predicting that the patient is a non-responder.
  • the medical information input unit 103 accepts input of medical information (predicted patient medical information) of a patient whose non-responder is not confirmed.
  • Predicted patient medical information includes elements similar to teacher data, such as patient attribute information.
  • the medical information input unit 103 can be configured with known hardware such as USB.
  • Patients who have not been confirmed to be non-responders include patients who have not been treated with CRT, but may also include patients who have been treated with CRT but cannot obtain a therapeutic effect. In the present specification, such a patient may be referred to as a "provisional non-responder".
  • the temporary non-responder may become a responder by changing the electrode arrangement of the CRT. Patients who do not obtain a therapeutic effect even if the CRT electrode arrangement is changed are confirmed non-responders.
  • the prediction unit 104 predicts whether or not the patient is a non-responder from the prediction target patient medical information input to the medical information input unit 103 based on the prediction determination model obtained by the machine learning unit 102.
  • the prediction result output unit 105 is a device that outputs the prediction result obtained by the prediction unit 104.
  • the prediction result output unit 105 can be configured by a known information output device such as a display controller that displays the prediction result on the display and a print controller that prints out the prediction result on the printer.
  • FIG. 2 shows a flowchart of the prediction process executed by the cardiac resynchronization therapy treatment effect prediction device 1.
  • a program for executing this process on the processor of the cardiac resynchronization therapy therapeutic effect prediction device 1 is stored in a computer-readable storage medium such as a memory or storage of the cardiac resynchronization therapy therapeutic effect prediction device 1.
  • step S1 the teacher data input unit 101 accepts the input of teacher data.
  • step S2 the machine learning unit 102 machine-learns a non-responder prediction determination model from the teacher data input to the teacher data input unit 101.
  • the machine learning unit 102 may also machine-learn a model for predicting and determining a temporary non-responder and a model for predicting and determining a change position of an electrode in which the temporary non-responder becomes a responder.
  • step S3 the medical information input unit 103 accepts the input of the prediction target patient medical information.
  • step S4 the prediction unit 104 determines whether or not the patient is a non-responder from the prediction target patient medical information input to the medical information input unit 103 based on the prediction determination model obtained by the machine learning unit 102. Is predicted and judged. Whether the patient is a tentative non-responder and the change position of the electrode where the tentative non-responder becomes a responder may be predicted and determined.
  • step S5 the prediction result output unit 105 outputs the result of the prediction determination obtained by the prediction unit 104.
  • the results of this prediction determination include whether the patient is predicted to be a non-responder, whether the patient is a non-responder but a provisional non-responder, and changes in the electrodes that make the provisional non-responder a responder. Position can be included.
  • cardiac resynchronization therapy therapeutic effect prediction device 1 By performing the above processing in the cardiac resynchronization therapy therapeutic effect prediction device 1, it is possible to predict patients who are not indicated for CRT, that is, non-responders, or patients who are indicated, that is, responders. As a result, CRT can be performed only on patients with a high therapeutic effect, wasteful physical, economic, and time burdens can be avoided, and the QOL of patients can be improved. Further, from the viewpoint of medical economics, it is possible to suppress the treatment of CRT on a non-responder having no therapeutic effect, and it is possible to reduce unnecessary medical expenses. In addition, patients who are predicted to be non-responders can be given the opportunity to propose alternative therapies such as artificial hearts, heart transplants, and regenerative medicine.
  • alternative therapies such as artificial hearts, heart transplants, and regenerative medicine.

Abstract

[Problème] Fournir un dispositif, un procédé et un programme grâce auxquels il est possible de prédire des non répondeurs à une thérapie de resynchronisation cardiaque (CRT) indépendamment de la gravité des symptômes d'insuffisance cardiaque. [Solution] À l'étape S1, une unité d'entrée de données d'apprentissage (101) reçoit une entrée de données d'apprentissage. À l'étape S2, une unité d'apprentissage machine (102) forme un modèle de détermination de prédiction de non répondeur à partir de l'entrée de données d'apprentissage vers l'unité d'entrée de données d'apprentissage. À l'étape S3, une unité d'entrée d'informations médicales (103) reçoit une entrée d'informations médicales d'un patient d'intérêt. À l'étape S4, une unité de prédiction (104) prédit et détermine si le patient est ou non un non répondeur à partir des informations médicales du patient d'intérêt entrées dans l'unité d'entrée d'informations médicales sur la base du modèle de détermination de prédiction obtenu par l'unité d'apprentissage machine. À l'étape S5, une unité de sortie de résultat de prédiction (105) délivre le résultat de prédiction et de détermination par l'unité de prédiction.
PCT/JP2020/038910 2019-11-07 2020-10-15 Dispositif, procédé et programme permettant de prédire l'effet thérapeutique d'une thérapie de resynchronisation cardiaque WO2021090659A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2019202536A JP2022183393A (ja) 2019-11-07 2019-11-07 心臓再同期療法治療効果予測装置、方法およびプログラム
JP2019-202536 2019-11-07

Publications (1)

Publication Number Publication Date
WO2021090659A1 true WO2021090659A1 (fr) 2021-05-14

Family

ID=75849948

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/038910 WO2021090659A1 (fr) 2019-11-07 2020-10-15 Dispositif, procédé et programme permettant de prédire l'effet thérapeutique d'une thérapie de resynchronisation cardiaque

Country Status (2)

Country Link
JP (1) JP2022183393A (fr)
WO (1) WO2021090659A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2806486C1 (ru) * 2022-11-23 2023-11-01 Федеральное государственное бюджетное учреждение науки институт иммунологии и физиологии Уральского отделения Российской академии наук Способ прогнозирования эффективности сердечной ресинхронизирующей терапии с использованием оптимизации расположения стимулирующих электродов

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010052303A1 (fr) * 2008-11-06 2010-05-14 Oslo Universitetssykehus Hf Analyse de données d'activation électromécanique ventriculaire
US20100280355A1 (en) * 2007-12-14 2010-11-04 Grimm Richard A System and method to characterize cardiac function
US20130072790A1 (en) * 2011-09-19 2013-03-21 University Of Pittsburgh-Of The Commonwealth System Of Higher Education Selection and optimization for cardiac resynchronization therapy
JP2013541385A (ja) * 2010-10-26 2013-11-14 オスロ ユニヴェルジテットサイケフス エイチエフ 心筋セグメントの仕事量分析方法
US10311978B2 (en) * 2012-01-30 2019-06-04 Siemens Healthcare Gmbh Method and system for patient specific planning of cardiac therapies on preoperative clinical data and medical images
WO2019208703A1 (fr) * 2018-04-26 2019-10-31 日本電気株式会社 Dispositif de traitement d'informations, procédé de commande et programme

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100280355A1 (en) * 2007-12-14 2010-11-04 Grimm Richard A System and method to characterize cardiac function
WO2010052303A1 (fr) * 2008-11-06 2010-05-14 Oslo Universitetssykehus Hf Analyse de données d'activation électromécanique ventriculaire
JP2013541385A (ja) * 2010-10-26 2013-11-14 オスロ ユニヴェルジテットサイケフス エイチエフ 心筋セグメントの仕事量分析方法
US20130072790A1 (en) * 2011-09-19 2013-03-21 University Of Pittsburgh-Of The Commonwealth System Of Higher Education Selection and optimization for cardiac resynchronization therapy
US10311978B2 (en) * 2012-01-30 2019-06-04 Siemens Healthcare Gmbh Method and system for patient specific planning of cardiac therapies on preoperative clinical data and medical images
WO2019208703A1 (fr) * 2018-04-26 2019-10-31 日本電気株式会社 Dispositif de traitement d'informations, procédé de commande et programme

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2806486C1 (ru) * 2022-11-23 2023-11-01 Федеральное государственное бюджетное учреждение науки институт иммунологии и физиологии Уральского отделения Российской академии наук Способ прогнозирования эффективности сердечной ресинхронизирующей терапии с использованием оптимизации расположения стимулирующих электродов

Also Published As

Publication number Publication date
JP2022183393A (ja) 2022-12-09

Similar Documents

Publication Publication Date Title
van der Bijl et al. Prognostic implications of global, left ventricular myocardial work efficiency before cardiac resynchronization therapy
Ypenburg et al. Optimal left ventricular lead position predicts reverse remodeling and survival after cardiac resynchronization therapy
US11398311B2 (en) System for visualizing heart activation
Shetty et al. Cardiac magnetic resonance-derived anatomy, scar, and dyssynchrony fused with fluoroscopy to guide LV lead placement in cardiac resynchronization therapy: a comparison with acute haemodynamic measures and echocardiographic reverse remodelling
Giraldi et al. Long-term effectiveness of cardiac resynchronization therapy in heart failure patients with unfavorable cardiac veins anatomy: comparison of surgical versus hemodynamic procedure
Schuler et al. Predictors of appropriate ICD therapy in patients with arrhythmogenic right ventricular cardiomyopathy: long term experience of a tertiary care center
Becker et al. Analysis of LV lead position in cardiac resynchronization therapy using different imaging modalities
Ramachandran et al. Singular value decomposition applied to cardiac strain from MR imaging for selection of optimal cardiac resynchronization therapy candidates
Truong et al. Utility of dual-source computed tomography in cardiac resynchronization therapy—DIRECT study
Naqvi Echocardiography-guided biventricular pacemaker optimization
Tomaske et al. Preserved cardiac synchrony and function with single-site left ventricular epicardial pacing during mid-term follow-up in paediatric patients
Zhang et al. Assessment of cardiac lead perforation: comparison among chest radiography, transthoracic echocardiography and electrocardiography-gated contrast-enhanced cardiac CT
Montenegro Camanho et al. Clinical outcomes and mortality in old and very old patients undergoing cardiac resynchronization therapy
WO2021090659A1 (fr) Dispositif, procédé et programme permettant de prédire l'effet thérapeutique d'une thérapie de resynchronisation cardiaque
Sarkar et al. Evaluation of long term effect of RV apical pacing on global LV function by echocardiography
Bilchick Does cardiac resynchronization therapy benefit patients with right bundle branch block: left ventricular free wall pacing: seldom right for right bundle branch block
Singh et al. Quantitative assessment of cardiac mechanical synchrony using equilibrium radionuclide angiography
Eitel et al. Baseline delayed left ventricular activation predicts long-term clinical outcome in cardiac resynchronization therapy recipients
Hutchins et al. Progressive left ventricular noncompaction and systolic dysfunction
Caloian et al. The influence of cardiac pacemaker programming modes on exercise capacity
Zhao et al. Computer-based assessment of ventricular mechanical synchrony from magnetic resonance imaging
D'IVERNOIS et al. Where are the left ventricular leads really implanted? A study of 90 consecutive patients
ElRefai et al. Role of deep learning methods in screening for subcutaneous implantable cardioverter defibrillator in heart failure
Ellenbogen et al. Delayed Response to Cardiac Resynchronization: Better Late Than Never
Auricchio et al. Status and update on cardiac resynchronization therapy trials

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20884990

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20884990

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP