Academic Journal
A Deep Learning-Based Automated Framework for Subpeak Designation on Intracranial Pressure Signals
| العنوان: | A Deep Learning-Based Automated Framework for Subpeak Designation on Intracranial Pressure Signals |
|---|---|
| المؤلفون: | Legé, Donatien, Gergelé, Laurent, Homme, Marion, Lapayre, Jean-Christophe, Launey, Yoann, Henriet, Julien |
| المساهمون: | Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) (FEMTO-ST), Université de Technologie de Belfort-Montbeliard (UTBM)-Ecole Nationale Supérieure de Mécanique et des Microtechniques (ENSMM)-Centre National de la Recherche Scientifique (CNRS)-Université de Franche-Comté (UFC), Université Bourgogne Franche-Comté COMUE (UBFC)-Université Bourgogne Franche-Comté COMUE (UBFC), Unité de soins intensifs médico-chirurgicaux CHU Saint-Etienne, Centre Hospitalier Universitaire de Saint-Etienne CHU Saint-Etienne (CHU ST-E), SOPHYSA (SOPHYSA), Université de Rennes (UR), Centre Hospitalier Universitaire de Rennes CHU Rennes = Rennes University Hospital Pontchaillou |
| المصدر: | Sensors ; https://hal.science/hal-04224624 ; Sensors, 2023, 23 (18), pp.7834 (17) |
| بيانات النشر: | HAL CCSD |
| سنة النشر: | 2023 |
| المجموعة: | Université de Franche-Comté (UFC): HAL |
| مصطلحات موضوعية: | [INFO.INFO-CR]Computer Science [cs]/Cryptography and Security [cs.CR], [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], [INFO.INFO-ET]Computer Science [cs]/Emerging Technologies [cs.ET], [INFO.INFO-IU]Computer Science [cs]/Ubiquitous Computing, [INFO.INFO-MA]Computer Science [cs]/Multiagent Systems [cs.MA], [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, [INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] |
| الوصف: | International audience ; The intracranial pressure (ICP) signal, as monitored on patients inintensive care units, contains pulses of cardiac origin, where P1and P2 subpeaks can often be observed. When calculable, the ratioof their relative amplitudes is an indicator of the patient’scerebral compliance. This characterization is particularlyinformative for the overall state of the cerebrospinal system. Theaim of this study is to develop and assess the performances of adeep learning-based pipeline for P2/P1 ratio computation that onlytakes a raw ICP signal as an input. The output P2/P1 ratio signalcan be discontinuous since P1 and P2 subpeaks are not alwaysvisible. The proposed pipeline performs four tasks, namely (i)heartbeat-induced pulse detection, (ii) pulse selection, (iii) P1and P2 designation, and (iv) signal smoothing and outlier removal.For tasks (i) and (ii), the performance of a recurrent neuralnetwork is compared to that of a convolutional neural network. Thefinal algorithm is evaluated on a 4344-pulse testing datasetsampled from 10 patient recordings. Pulse selection is achievedwith an area under the curve of 0.90, whereas the subpeakdesignation algorithm identifies pulses with a P2/P1 ratio > 1with 97.3% accuracy. Although it still needs to be evaluated on alarger number of labeled recordings, our automated P2/P1 ratiocalculation framework appears to be a promising tool that can beeasily embedded into bedside monitoring devices. |
| نوع الوثيقة: | article in journal/newspaper |
| اللغة: | English |
| Relation: | hal-04224624; https://hal.science/hal-04224624; https://hal.science/hal-04224624/document; https://hal.science/hal-04224624/file/adfb3b56-4cce-456d-82fa-533c75a3259e-author.pdf |
| الاتاحة: | https://hal.science/hal-04224624 https://hal.science/hal-04224624/document https://hal.science/hal-04224624/file/adfb3b56-4cce-456d-82fa-533c75a3259e-author.pdf |
| Rights: | info:eu-repo/semantics/OpenAccess |
| رقم الانضمام: | edsbas.1CF357A6 |
| قاعدة البيانات: | BASE |
| الوصف غير متاح. |