Jaryd Hunter

Jaryd Hunter

Auto-generated profile, based on publicly available information

Jaryd Hunter

DATA Team, University Health Network, Toronto, Canada

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 September 2024 in npj Digital Medicine
Verified
Journal Article
Open access
PDF file
SPSergey Protserov
Sergey Protserov
JHJaryd Hunter
Jaryd Hunter
HZHaochi Zhang
Haochi Zhang
PMPouria Mashouri
Pouria Mashouri
CMCaterina Masino
Caterina Masino
AMAmin Madani
Amin Madani
Deep learning for computer vision can be leveraged for interpreting surgical scenes and providing surgeons with real-time guidance to avoid complications. However, neither generalizability nor scalability of computer-vision-based surgical guidance systems have been demonstrated, especially to geographic locations that lack hardware and infrastructure necessary for real-time inference. We propose a new equipment-agnostic framework for real-time use in operating suites. Using laparoscopic cholecystectomy and semantic segmentation models for predicting safe/dangerous (“Go”/”No-Go”) zones of dissection as an example use case, this study aimed to develop and test the performance of a novel data pipeline linked to a web-platform that enables real-time deployment from any edge device. To test this infrastructure and demonstrate its scalability and generalizability, lightweight U-Net and SegFormer models were trained on annotated frames from a large and diverse multicenter dataset from 136 institutions, and then tested on a separate prospectively collected dataset. A web-platform was created to enable real-time inference on any surgical video stream, and performance was tested on and optimized for a range of network speeds. The U-Net and SegFormer models respectively achieved mean Dice scores of 57% and 60%, precision 45% and 53%, and recall 82% and 75% for predicting the Go zone, and mean Dice scores of 76% and 76%, precision 68% and 68%, and recall 92% and 92% for predicting the No-Go zone. After optimization of the client-server interaction over the network, we deliver a prediction stream of at least 60 fps and with a maximum round-trip delay of 70 ms for speeds above 8 Mbps. Clinical deployment of machine learning models for surgical guidance is feasible and cost-effective using a generalizable, scalable and equipment-agnostic framework that lacks dependency on hardware with high computing performance or ultra-fast internet connection speed.

18 May 2022 in Nature Communications
Verified
Journal Article
Open access
PDF file
TKTom Kaufman
Tom Kaufman
ENErez Nitzan
Erez Nitzan
NFNir Firestein
Nir Firestein
MGMiriam Bracha Ginzberg
Miriam Bracha Ginzberg
SISeshu Iyengar
Seshu Iyengar
RSRavid Straussman
Ravid Straussman
While multiplexing samples using DNA barcoding revolutionized the pace of biomedical discovery, multiplexing of live imaging-based applications has been limited by the number of fluorescent proteins that can be deconvoluted using common microscopy equipment. To address this limitation, we develop visual barcodes that discriminate the clonal identity of single cells by different fluorescent proteins that are targeted to specific subcellular locations. We demonstrate that deconvolution of these barcodes is highly accurate and robust to many cellular perturbations. We then use visual barcodes to generate ‘Signalome’ cell-lines by mixing 12 clones of different live reporters into a single population, allowing simultaneous monitoring of the activity in 12 branches of signaling, at clonal resolution, over time. Using the ‘Signalome’ we identify two distinct clusters of signaling pathways that balance growth and proliferation, emphasizing the importance of growth homeostasis as a central organizing principle in cancer signaling. The ability to multiplex samples in live imaging applications, both in vitro and in vivo may allow better high-content characterization of complex biological systems.

25 July 2024 in JMIR Formative Research
Verified
Journal Article
Open access
MNMohammad Noroozi
Mohammad Noroozi
AJAce St John
Ace St John
CMCaterina Masino
Caterina Masino
SLSimon Laplante
Simon Laplante
JHJaryd Hunter
Jaryd Hunter
MKMarta Kersten-Oertel
Marta Kersten-Oertel
Background: Major bile duct injuries during laparoscopic cholecystectomy (LC), often stemming from errors in surgical judgment and visual misperception of critical anatomy, significantly impact morbidity, mortality, disability, and health care costs. Objective: To enhance safe LC learning, we developed an educational mobile game, LapBot Safe Chole, which uses an artificial intelligence (AI) model to provide real-time coaching and feedback, improving intraoperative decision-making. Methods: LapBot Safe Chole offers a free, accessible simulated learning experience with real-time AI feedback. Players engage with intraoperative LC scenarios (short video clips) and identify ideal dissection zones. After the response, users receive an accuracy score from a validated AI algorithm. The game consists of 5 levels of increasing difficulty based on the Parkland grading scale for cholecystitis. Results: Beta testing (n=29) showed score improvements with each round, with attendings and senior trainees achieving top scores faster than junior residents. Learning curves and progression distinguished candidates, with a significant association between user level and scores (P=.003). Players found LapBot enjoyable and educational. Conclusions: LapBot Safe Chole effectively integrates safe LC principles into a fun, accessible, and educational game using AI-generated feedback. Initial beta testing supports the validity of the assessment scores and suggests high adoption and engagement potential among surgical trainees.

18 September 2023 in JMIR Preprints
Verified
Preprint
Open access
MNMohammad Noroozi
Mohammad Noroozi
AJAce St John
Ace St John
CMCaterina Masino
Caterina Masino
SLSimon Laplante
Simon Laplante
JHJaryd Hunter
Jaryd Hunter
MKMarta Kersten-Oertel
Marta Kersten-Oertel
BACKGROUND Major bile duct injuries during laparoscopic cholecystectomy (LC), often stemming from errors in surgical judgment and visual misperception of critical anatomy, significantly impact morbidity, mortality, disability, and health care costs. OBJECTIVE To enhance safe LC learning, we developed an educational mobile game, LapBot Safe Chole, which uses an artificial intelligence (AI) model to provide real-time coaching and feedback, improving intraoperative decision-making. METHODS LapBot Safe Chole offers a free, accessible simulated learning experience with real-time AI feedback. Players engage with intraoperative LC scenarios (short video clips) and identify ideal dissection zones. After the response, users receive an accuracy score from a validated AI algorithm. The game consists of 5 levels of increasing difficulty based on the Parkland grading scale for cholecystitis. RESULTS Beta testing (n=29) showed score improvements with each round, with attendings and senior trainees achieving top scores faster than junior residents. Learning curves and progression distinguished candidates, with a significant association between user level and scores (P=.003). Players found LapBot enjoyable and educational. CONCLUSIONS LapBot Safe Chole effectively integrates safe LC principles into a fun, accessible, and educational game using AI-generated feedback. Initial beta testing supports the validity of the assessment scores and suggests high adoption and engagement potential among surgical trainees.