Hospitalization prediction from the emergency department using computer vision AI with short patient video clips

被引：0

作者：

Ip, Wui ^{[1
]}

Xenochristou, Maria ^{[2
]}

Sui, Elaine ^{[3
]}

Ruan, Elyse ^{[4
]}

Ribeira, Ryan ^{[5
]}

Dash, Debadutta ^{[5
]}

Srinivasan, Malathi ^{[6
]}

Artandi, Maja ^{[6
]}

Omiye, Jesutofunmi A. ^{[2
,7
]}

Scoulios, Nicholas ^{[6
]}

Hofmann, Hayden L. ^{[2
]}

Mottaghi, Ali ^{[8
]}

Weng, Zhenzhen ^{[9
]}

Kumar, Abhinav ^{[3
]}

Ganesh, Ananya ^{[3
]}

Fries, Jason ^{[10
]}

Yeung-Levy, Serena ^{[2
,3
,8
,11
,12
]}

Hofmann, Lawrence V. ^{[4
,13
]}

机构：

[1] Stanford Univ, Sch Med, Dept Pediat, Palo Alto, CA 94303 USA

[2] Stanford Univ, Sch Med, Dept Biomed Data Sci, Palo Alto, CA USA

[3] Stanford Univ, Dept Comp Sci, Palo Alto, CA USA

[4] Stanford Hlth Care, Digital Hlth Care Integrat, Palo Alto, CA USA

[5] Stanford Univ, Sch Med, Dept Emergency Med, Palo Alto, CA USA

[6] Stanford Univ, Dept Med, Sch Med, Palo Alto, CA USA

[7] Stanford Univ, Dept Dermatol, Sch Med, Palo Alto, CA USA

[8] Stanford Univ, Dept Elect Engn, Palo Alto, CA USA

[9] Stanford Univ, Inst Computat & Math Engn, Palo Alto, CA USA

[10] Stanford Ctr Biomed Informat Res, Palo Alto, CA USA

[11] Stanford Univ, Clin Excellence Res Ctr, Sch Med, Palo Alto, CA USA

[12] Chan Zuckerberg Biohub San Francisco, San Francisco, CA USA

[13] Stanford Univ, Dept Radiol, Sch Med, Palo Alto, CA USA

来源：

NPJ DIGITAL MEDICINE | 2024年 / 7卷 / 01期

关键词：

SYSTEM; 1; DISPOSITION; ADMISSION; VARIABILITY; ACCURACY; DECISION; ACUITY; MODELS; TRIAGE;

D O I：

10.1038/s41746-024-01375-3

中图分类号：

R19 [保健组织与事业（卫生事业管理）];

学科分类号：

摘要：

In this study, we investigate the performance of computer vision AI algorithms in predicting patient disposition from the emergency department (ED) using short video clips. Clinicians often use "eye-balling" or clinical gestalt to aid in triage, based on brief observations. We hypothesize that AI can similarly use patient appearance for disposition prediction. Data were collected from adult patients at an academic ED, with mobile phone videos capturing patients performing simple tasks. Our AI algorithm, using video alone, showed better performance in predicting hospital admissions (AUROC = 0.693 [95% CI 0.689, 0.696]) compared to models using triage clinical data (AUROC = 0.678 [95% CI 0.668, 0.687]). Combining video and triage data achieved the highest predictive performance (AUROC = 0.714 [95% CI 0.709, 0.719]). This study demonstrates the potential of video AI algorithms to support ED triage and alleviate healthcare capacity strains during periods of high demand.

引用

页数：8

共 50 条

[21] Pilot Study of Emergency Department Patient Vital Signs Acquisition Using Experimental Video Photoplethysmography and Passive Infrared Thermography Devices
Kobayashi, Leo
Chuck, Carlin C.
Kim, Chris K.
Luchette, Katherine R.
Oster, Alana
Merck, Derek L.
Kirenko, Ihor
Van Zon, Kees
Bartula, Marek
Rocque, Mukul
Wang, Haibo
Capraro, Geoffrey A.
2019 IEEE 10TH ANNUAL UBIQUITOUS COMPUTING, ELECTRONICS & MOBILE COMMUNICATION CONFERENCE (UEMCON), 2019, : 23 - 32
[22] Odds of return: a prospective study using provider assessment to predict short-term patient return visits to the emergency department
Fruhan, Scott
Bills, Corey B.
BMJ OPEN, 2021, 11 (12):
[23] Predicting Patient Disposition from the Emergency Department: A Prospective Comparison of Attending Physician Gestalt vs. an Automated Computer Model
Barak-Corren, Y.
Agarwal, I
Reis, B. Y.
Fine, A. M.
ANNALS OF EMERGENCY MEDICINE, 2019, 74 (04) : S51 - S51
[24] Improving Patient Trajectory Forecasts in Hospitals: Using Emergency Department Data for Length of Stay Prediction and Next Hospital Unit Classification
Winter, Alexander
Kirsten, Toralf
Hartwig, Mattis
BIOMEDICAL ENGINEERING SYSTEMS AND TECHNOLOGIES, BIOSTEC 2023, 2024, 2079 : 84 - 106
[25] Redesigning the Information Highway - Using Video to Improve Nurse Satisfaction with Patients' Transitions from the Emergency Department to the Neuroscience Unit
Richardson, Melissa
Salehi, Omid
Hinkelman, Kevin
Tyan, Clarke
Young, Teresa
Meyers, Melissa
Holbert, Lori
Vandergriff, Melissa
STROKE, 2016, 47
[26] Assessing time to treatment and patient inflow in a Danish emergency department: A cohort study using data from electronic emergency screen boards
Nielsen R.F.
Pérez N.
Petersen P.
Biering K.
BMC Research Notes, 7 (1)
[27] Early short-term prediction of emergency department length of stay using natural language processing for low-acuity outpatients
Chen, Chien-Hua
Hsieh, Jer-Guang
Cheng, Shu-Ling
Lin, Yih-Lon
Lin, Po-Hsiang
Jeng, Jyh-Horng
AMERICAN JOURNAL OF EMERGENCY MEDICINE, 2020, 38 (11): : 2368 - 2373
[28] Advances in estimating pedestrian measures through artificial intelligence: From data sources, computer vision, video analytics to the prediction of crash frequency
Lian, Ting
Loo, Becky P. Y.
Fan, Zhuangyuan
COMPUTERS ENVIRONMENT AND URBAN SYSTEMS, 2024, 107
[29] Derivation and Validation of a Hospital Admission Prediction Model Adding Coded Chief Complaint to Demographic, Emergency Department Operational and Patient Acuity Data Available at Emergency Department Triage Using Neural Net Methodology
Handly, N.
Thompson, D. A.
Venkat, A.
ANNALS OF EMERGENCY MEDICINE, 2013, 62 (04) : S138 - S138
[30] Noninvasive ML-based BUN Prediction from Ten-second Fingertip Video using Generative AI
Sridevi, Parama
Arefin, Kazi Zawad
Das, Dipranjan
Khan, Rumi Ahmed
Ahamed, Sheikh Iqbal
2024 IEEE 48TH ANNUAL COMPUTERS, SOFTWARE, AND APPLICATIONS CONFERENCE, COMPSAC 2024, 2024, : 1630 - 1635

← 1 2 3 4 5 →