Jacob Sunshine

Jacob Sunshine

Auto-generated profile, based on publicly available information

Jacob Sunshine

Google Research, Seattle, WA, USA.

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

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

Representative Publications

View All Publications
9 April 2025 in Nature
Verified
Journal Article
Open access
DMDaniel McDuff
Daniel McDuff
MSMike Schaekermann
Mike Schaekermann
TTTao Tu
Tao Tu
APAnil Palepu
Anil Palepu
AWAmy Wang
Amy Wang
VNVivek Natarajan
Vivek Natarajan
A comprehensive differential diagnosis is a cornerstone of medical care that is often reached through an iterative process of interpretation that combines clinical history, physical examination, investigations and procedures. Interactive interfaces powered by large language models present new opportunities to assist and automate aspects of this process1. Here we introduce the Articulate Medical Intelligence Explorer (AMIE), a large language model that is optimized for diagnostic reasoning, and evaluate its ability to generate a differential diagnosis alone or as an aid to clinicians. Twenty clinicians evaluated 302 challenging, real-world medical cases sourced from published case reports. Each case report was read by two clinicians, who were randomized to one of two assistive conditions: assistance from search engines and standard medical resources; or assistance from AMIE in addition to these tools. All clinicians provided a baseline, unassisted differential diagnosis prior to using the respective assistive tools. AMIE exhibited standalone performance that exceeded that of unassisted clinicians (top-10 accuracy 59.1% versus 33.6%, P = 0.04). Comparing the two assisted study arms, the differential diagnosis quality score was higher for clinicians assisted by AMIE (top-10 accuracy 51.7%) compared with clinicians without its assistance (36.1%; McNemar’s test: 45.7, P < 0.01) and clinicians with search (44.4%; McNemar’s test: 4.75, P = 0.03). Further, clinicians assisted by AMIE arrived at more comprehensive differential lists than those without assistance from AMIE. Our study suggests that AMIE has potential to improve clinicians’ diagnostic reasoning and accuracy in challenging cases, meriting further real-world evaluation for its ability to empower physicians and widen patients’ access to specialist-level expertise.

19 June 2019 in npj Digital Medicine
Verified
Journal Article
Open access
PDF file
JCJustin Chan
Justin Chan
TRThomas Rea
Thomas Rea
SGShyamnath Gollakota
Shyamnath Gollakota
JSJacob E. Sunshine
Jacob E. Sunshine
Out-of-hospital cardiac arrest is a leading cause of death worldwide. Rapid diagnosis and initiation of cardiopulmonary resuscitation (CPR) is the cornerstone of therapy for victims of cardiac arrest. Yet a significant fraction of cardiac arrest victims have no chance of survival because they experience an unwitnessed event, often in the privacy of their own homes. An under-appreciated diagnostic element of cardiac arrest is the presence of agonal breathing, an audible biomarker and brainstem reflex that arises in the setting of severe hypoxia. Here, we demonstrate that a support vector machine (SVM) can classify agonal breathing instances in real-time within a bedroom environment. Using real-world labeled 9-1-1 audio of cardiac arrests, we train the SVM to accurately classify agonal breathing instances. We obtain an area under the curve (AUC) of 0.9993 ± 0.0003 and an operating point with an overall sensitivity and specificity of 97.24% (95% CI: 96.86–97.61%) and 99.51% (95% CI: 99.35–99.67%). We achieve a false positive rate between 0 and 0.14% over 82 h (117,985 audio segments) of polysomnographic sleep lab data that includes snoring, hypopnea, central, and obstructive sleep apnea events. We also evaluate our classifier in home sleep environments: the false positive rate was 0–0.22% over 164 h (236,666 audio segments) of sleep data collected across 35 different bedroom environments. We prototype our proof-of-concept contactless system using commodity smart devices (Amazon Echo and Apple iPhone) and demonstrate its effectiveness in identifying cardiac arrest-associated agonal breathing instances played over the air.

9 January 2019 in Science Translational Medicine
Verified
Journal Article
Open access
PDF file
RNRajalakshmi Nandakumar
Rajalakshmi Nandakumar
SGShyamnath Gollakota
Shyamnath Gollakota
JSJacob E. Sunshine
Jacob E. Sunshine
Early detection and rapid intervention can prevent death from opioid overdose. At high doses, opioids (particularly fentanyl) can cause rapid cessation of breathing (apnea), hypoxemic/hypercarbic respiratory failure, and death, the physiologic sequence by which people commonly succumb from unintentional opioid overdose. We present algorithms that run on smartphones and unobtrusively detect opioid overdose events and their precursors. Our proof-of- concept contactless system converts the phone into a short-range active sonar using frequency shifts to identify respiratory depression, apnea, and gross motor movements associated with acute opioid toxicity. We develop algorithms and perform testing in two environments: (i) an approved supervised injection facility (SIF), where people self-inject illicit opioids, and (ii) the operating room (OR), where we simulate rapid, opioid-induced overdose events using routine induction of general anesthesia. In the SIF (n = 209), our system identified postinjection, opioid-induced central apnea with 96% sensitivity and 98% specificity and identified respiratory depression with 87% sensitivity and 89% specificity. These two key events commonly precede fatal opioid overdose. In the OR, our algorithm identified 19 of 20 simulated overdose events. Given the reliable reversibility of acute opioid toxicity, smartphone-enabled overdose detection coupled with the ability to alert naloxone-equipped friends and family or emergency medical services (EMS) could hold potential as a low-barrier, harm reduction intervention.

1 April 2013 in Respiratory Care
Verified
Journal Article
Closed access
JSJacob E Sunshine
Jacob E Sunshine
SDSteven Deem
Steven Deem
NWNoel S Weiss
Noel S Weiss
NYN David Yanez
N David Yanez
SDStephen Daniel
Stephen Daniel
MTMiriam M Treggiari
Miriam M Treggiari
In critically ill patients, induction with etomidate is hypothesized to be associated with an increased risk of mortality. Previous randomized studies suggest a modest trend toward an increased risk of death among etomidate recipients; however, this relationship has not been measured with great statistical precision. We aimed to test whether etomidate is associated with risk of hospital mortality and other clinical outcomes in critically ill patients. We conducted a retrospective cohort study from January 1, 2001, to December 31, 2005, of 824 subjects requiring mechanical ventilation, who underwent adrenal function testing in the ICUs of 2 academic medical centers. The primary outcome was in-hospital mortality, comparing subjects given etomidate (n = 452) to those given an alternative induction agent (n = 372). The secondary outcome was diagnosis of critical illness-related corticosteroid insufficiency following etomidate exposure. Overall mortality was 34.3%. After adjustment for age, sex, and baseline illness severity, the relative risk of death among the etomidate recipients was higher than that of subjects given an alternative agent (relative risk 1.20, 95% CI 0.99-1.45). Among subjects whose adrenal function was assessed within the 48 hours following intubation, the adjusted risk of meeting the criteria for critical illness-related corticosteroid insufficiency was 1.37 (95% CI 1.12-1.66), comparing etomidate recipients to subjects given another induction agent. In this study of critically ill patients requiring endotracheal intubation, etomidate administration was associated with a trend toward a relative increase in mortality, similar to the collective results of smaller randomized trials conducted to date. If a small relative increased risk is truly present, though previous trials have been underpowered to detect it, in absolute terms the number of deaths associated with etomidate in this high-risk population would be considerable. Large, prospective controlled trials are needed to finalize the role of etomidate in critically ill patients.

26 February 2025 in Nature
Verified
Journal Article
Open access
PDF file
KSKamal Shah
Kamal Shah
AWAnran Wang
Anran Wang
YCYiwen Chen
Yiwen Chen
JMJitender Munjal
Jitender Munjal
SCSumeet Chhabra
Sumeet Chhabra
JSJake Sunshine
Jake Sunshine
Out-of-hospital cardiac arrest is a time-sensitive emergency that requires prompt identification and intervention: sudden, unwitnessed cardiac arrest is nearly unsurvivable1–3. A cardinal sign of cardiac arrest is sudden loss of pulse4. Automated biosensor detection of unwitnessed cardiac arrest, and dispatch of medical assistance, may improve survivability given the significant prognostic role of time3,5, but only if the false positive burden on public emergency medical systems is minimized5–7. Here we show that a multimodal, machine learning-based algorithm on a smartwatch can reach performance thresholds making it deployable at societal scale. First, using photoplethysmography (PPG), we show that wearable PPG measurements of peripheral pulselessness (induced via an arterial occlusion model) manifest similarly to pulselessness caused by a common cardiac arrest arrhythmia, ventricular fibrillation (VF). Based on the similarity of the PPG signal (from VF or arterial occlusion), we developed and validated a loss of pulse detection algorithm using data from peripheral pulselessness and free-living conditions. Once developed, we evaluated the end-to-end algorithm prospectively: there was 1 unintentional emergency call per 21.67 user-years across two prospective studies; the sensitivity was 67.23% (95% confidence interval, 64.32%–70.05%) in a prospective arterial occlusion cardiac arrest simulation model. These results suggest a new opportunity, deployable at scale, for wearable-based detection of sudden loss of pulse while minimizing societal costs of excess false detections7.

22 November 2021 in Scientific Reports
Verified
Journal Article
Open access
PDF file
JCJustin Chan
Justin Chan
VIVikram Iyer
Vikram Iyer
AWAnran Wang
Anran Wang
ALAlexander Lyness
Alexander Lyness
PKPreetma Kooner
Preetma Kooner
SGShyamnath Gollakota
Shyamnath Gollakota
Overdoses from non-medical use of opioids can lead to hypoxemic/hypercarbic respiratory failure, cardiac arrest, and death when left untreated. Opioid toxicity is readily reversed with naloxone, a competitive antagonist that can restore respiration. However, there remains a critical need for technologies to administer naloxone in the event of unwitnessed overdose events. We report a closed-loop wearable injector system that measures respiration and apneic motion associated with an opioid overdose event using a pair of on-body accelerometers, and administers naloxone subcutaneously upon detection of an apnea. Our proof-of-concept system has been evaluated in two environments: (i) an approved supervised injection facility (SIF) where people self-inject opioids under medical supervision and (ii) a hospital environment where we simulate opioid-induced apneas in healthy participants. In the SIF (n = 25), our system identified breathing rate and post-injection respiratory depression accurately when compared to a respiratory belt. In the hospital, our algorithm identified simulated apneic events and successfully injected participants with 1.2 mg of naloxone. Naloxone delivery was verified by intravenous blood draw post-injection for all participants. A closed-loop naloxone injector system has the potential to complement existing evidence-based harm reduction strategies and, in the absence of bystanders, help make opioid toxicity events functionally witnessed and in turn more likely to be successfully resuscitated.