CaltechAUTHORS
  A Caltech Library Service

Deep Neural Networks Can Accurately Detect Blood Loss and Hemorrhage Control Task Success From Video

Kugener, Guillaume and Zhu, Yichao and Pangal, Dhiraj J. and Sinha, Aditya and Markarian, Nicholas and Roshannai, Arman and Chan, Justin and Anandkumar, Animashree and Hung, Andrew J. and Wrobel, Bozena B. and Zada, Gabriel and Donoho, Daniel A. (2022) Deep Neural Networks Can Accurately Detect Blood Loss and Hemorrhage Control Task Success From Video. Neurosurgery, 90 (6). pp. 823-829. ISSN 0148-396X. doi:10.1227/neu.0000000000001906. https://resolver.caltech.edu/CaltechAUTHORS:20220413-607067100

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20220413-607067100

Abstract

Background: Deep neural networks (DNNs) have not been proven to detect blood loss (BL) or predict surgeon performance from video. Objective: To train a DNN using video from cadaveric training exercises of surgeons controlling simulated internal carotid hemorrhage to predict clinically relevant outcomes. Methods: Video was input as a series of images; deep learning networks were developed, which predicted BL and task success from images alone (automated model) and images plus human-labeled instrument annotations (semiautomated model). These models were compared against 2 reference models, which used average BL across all trials as its prediction (control 1) and a linear regression with time to hemostasis (a metric with known association with BL) as input (control 2). The root-mean-square error (RMSE) and correlation coefficients were used to compare the models; lower RMSE indicates superior performance. Results: One hundred forty-three trials were used (123 for training and 20 for testing). Deep learning models outperformed controls (control 1: RMSE 489 mL, control 2: RMSE 431 mL, R2 = 0.35) at BL prediction. The automated model predicted BL with an RMSE of 358 mL (R2 = 0.4) and correctly classified outcome in 85% of trials. The RMSE and classification performance of the semiautomated model improved to 260 mL and 90%, respectively. Conclusion: BL and task outcome classification are important components of an automated assessment of surgical performance. DNNs can predict BL and outcome of hemorrhage control from video alone; their performance is improved with surgical instrument presence data. The generalizability of DNNs trained on hemorrhage control tasks should be investigated.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1227/neu.0000000000001906DOIArticle
ORCID:
AuthorORCID
Kugener, Guillaume0000-0002-4697-2847
Pangal, Dhiraj J.0000-0001-7391-9825
Anandkumar, Animashree0000-0002-6974-6797
Zada, Gabriel0000-0001-5821-902X
Donoho, Daniel A.0000-0002-0531-1436
Additional Information:© 2022 Congress of Neurological Surgeons.
Subject Keywords:Artificial intelligence, Complication management, Hemorrhage control, Machine learning, Video
Issue or Number:6
DOI:10.1227/neu.0000000000001906
Record Number:CaltechAUTHORS:20220413-607067100
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20220413-607067100
Official Citation:Kugener, Guillaume MEng; Zhu, Yichao MS; Pangal, Dhiraj J. BS; Sinha, Aditya BS; Markarian, Nicholas BS; Roshannai, Arman; Chan, Justin BS; Anandkumar, Animashree PhD; Hung, Andrew J. MD; Wrobel, Bozena B. MD; Zada, Gabriel MD, MS; Donoho, Daniel A. MD. Deep Neural Networks Can Accurately Detect Blood Loss and Hemorrhage Control Task Success From Video, Neurosurgery: June 2022 - Volume 90 - Issue 6 - p 823-829 doi: 10.1227/neu.0000000000001906
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:114284
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:13 Apr 2022 21:38
Last Modified:27 Jul 2022 20:07

Repository Staff Only: item control page