Honors & Awards
Howard Hughes Medical Institute Medical Research Fellow, HHMI (2015-2016)
MD, Washington University in St. Louis School of Medicine (2018)
BS, California Institute of Technology, Bioengineering (2012)
Efficient Augmented Intelligence Framework for Bladder Lesion Detection.
JCO clinical cancer informatics
2023; 7: e2300031
Development of intelligence systems for bladder lesion detection is cost intensive. An efficient strategy to develop such intelligence solutions is needed.We used four deep learning models (ConvNeXt, PlexusNet, MobileNet, and SwinTransformer) covering a variety of model complexity and efficacy. We trained these models on a previously published educational cystoscopy atlas (n = 312 images) to estimate the ratio between normal and cancer scores and externally validated on cystoscopy videos from 68 cases, with region of interest (ROI) pathologically confirmed to be benign and cancerous bladder lesions (ie, ROI). The performance measurement included specificity and sensitivity at frame level, frame sequence (block) level, and ROI level for each case.Specificity was comparable between four models at frame (range, 30.0%-44.8%) and block levels (56%-67%). Although sensitivity at the frame level (range, 81.4%-88.1%) differed between the models, sensitivity at the block level (100%) and ROI level (100%) was comparable between these models. MobileNet and PlexusNet were computationally more efficient for real-time ROI detection than ConvNeXt and SwinTransformer.Educational cystoscopy atlas and efficient models facilitate the development of real-time intelligence system for bladder lesion detection.
View details for DOI 10.1200/CCI.23.00031
View details for PubMedID 37774313
A magnetic hydrogel for the efficient retrieval of kidney stone fragments during ureteroscopy.
2023; 14 (1): 3711
Only 60-75% of conventional kidney stone surgeries achieve complete stone-free status. Up to 30% of patients with residual fragments <2 mm in size experience subsequent stone-related complications. Here we demonstrate a stone retrieval technology in which fragments are rendered magnetizable with a magnetic hydrogel so that they can be easily retrieved with a simple magnetic tool. The magnetic hydrogel facilitates robust in vitro capture of stone fragments of clinically relevant sizes and compositions. The hydrogel components exhibit no cytotoxicity in cell culture and only superficial effects on ex vivo human urothelium and in vivo mouse bladders. Furthermore, the hydrogel demonstrates antimicrobial activity against common uropathogens on par with that of common antibiotics. By enabling the efficient retrieval of kidney stone fragments, our method can lead to improved stone-free rates and patient outcomes.
View details for DOI 10.1038/s41467-023-38936-1
View details for PubMedID 37349287
View details for PubMedCentralID 5853829
Conceptual Framework and Documentation Standards of Cystoscopic Media Content for Artificial Intelligence.
Journal of biomedical informatics
The clinical documentation of cystoscopy includes visual and textual materials. However, the secondary use of visual cystoscopic data for educational and research purposes remains limited due to inefficient data management in routine clinical practice.A conceptual framework was designed to document cystoscopy in a standardized manner with three major sections: data management, annotation management, and utilization management. A Swiss-cheese model was proposed for quality control and root cause analyses. We defined the infrastructure required to implement the framework with respect to FAIR (findable, accessible, interoperable, reusable) principles. We applied two scenarios exemplifying data sharing for research and educational projects to ensure compliance with FAIR principles.The framework was successfully implemented while following FAIR principles. The cystoscopy atlas produced from the framework could be presented in an educational web portal; a total of 68 full-length qualitative videos and corresponding annotation data were sharable for artificial intelligence projects covering frame classification and segmentation problems at case, lesion, and frame levels.Our study shows that the proposed framework facilitates the storage of visual documentation in a standardized manner and enables FAIR data for education and artificial intelligence research.
View details for DOI 10.1016/j.jbi.2023.104369
View details for PubMedID 37088456
- Passive monitoring by smart toilets for precision health. Science translational medicine 2023; 15 (681): eabk3489
Conceptual Framework and Documentation Standards of Cystoscopic Media Content for Artificial Intelligence.
The clinical documentation of cystoscopy includes visual and textual materials. However, the secondary use of visual cystoscopic data for educational and research purposes remains limited due to inefficient data management in routine clinical practice.A conceptual framework was designed to document cystoscopy in a standardized manner with three major sections: data management, annotation management, and utilization management. A Swiss-cheese model was proposed for quality control and root cause analyses. We defined the infrastructure required to implement the framework with respect to FAIR (findable, accessible, interoperable, re-usable) principles. We applied two scenarios exemplifying data sharing for research and educational projects to ensure the compliance with FAIR principles.The framework was successfully implemented while following FAIR principles. The cystoscopy atlas produced from the framework could be presented in an educational web portal; a total of 68 full-length qualitative videos and corresponding annotation data were sharable for artificial intelligence projects covering frame classification and segmentation problems at case, lesion and frame levels.Our study shows that the proposed framework facilitates the storage of the visual documentation in a standardized manner and enables FAIR data for education and artificial intelligence research.
View details for DOI 10.2196/32800
View details for PubMedID 36713258
View details for PubMedCentralID PMC9882574
An Efficient Framework for Video Documentation of Bladder Lesions for Cystoscopy: A Proof-of-Concept Study.
Journal of medical systems
2022; 46 (11): 73
Processing full-length cystoscopy videos is challenging for documentation and research purposes. We therefore designed a surgeon-guided framework to extract short video clips with bladder lesions for more efficient content navigation and extraction. Screenshots of bladder lesions were captured during transurethral resection of bladder tumor, then manually labeled according to case identification, date, lesion location, imaging modality, and pathology. The framework used the screenshot to search for and extract a corresponding 10-seconds video clip. Each video clip included a one-second space holder with a QR barcode informing the video content. The success of the framework was measured by the secondary use of these short clips and the reduction of storage volume required for video materials. From 86 cases, the framework successfully generated 249 video clips from 230 screenshots, with 14 erroneous video clips from 8 screenshots excluded. The HIPPA-compliant barcodes provided information of video contents with a 100% data completeness. A web-based educational gallery was curated with various diagnostic categories and annotated frame sequences. Compared with the unedited videos, the informative short video clips reduced the storage volume by 99.5%. In conclusion, our framework expedites the generation of visual contents with surgeon's instruction for cystoscopy and potential incorporation of video data towards applications including clinical documentation, education, and research.
View details for DOI 10.1007/s10916-022-01862-8
View details for PubMedID 36190581
- Resident and Fellow Unions: Collective Activism to Promote Well-being for Physicians in Training. JAMA 2022
- Smart toilets for monitoring COVID-19 surges: passive diagnostics and public health. NPJ digital medicine 2022; 5 (1): 39
Continuous health monitoring: An opportunity for precision health.
Science translational medicine
2021; 13 (597)
Continuous health monitoring and integrated diagnostic devices, worn on the body and used in the home, will help to identify and prevent early manifestations of disease. However, challenges lie ahead in validating new health monitoring technologies and in optimizing data analytics to extract actionable conclusions from continuously obtained health data.
View details for DOI 10.1126/scitranslmed.abe5383
View details for PubMedID 34108250
- Digital biomarkers in human excreta. Nature reviews. Gastroenterology & hepatology 2021
A mountable toilet system for personalized health monitoring via the analysis of excreta.
Nature biomedical engineering
Technologies for the longitudinal monitoring of a person's health are poorly integrated with clinical workflows, and have rarely produced actionable biometric data for healthcare providers. Here, we describe easily deployable hardware and software for the long-term analysis of a user's excreta through data collection and models of human health. The 'smart' toilet, which is self-contained and operates autonomously by leveraging pressure and motion sensors, analyses the user's urine using a standard-of-care colorimetric assay that traces red-green-blue values from images of urinalysis strips, calculates the flow rate and volume of urine using computer vision as a uroflowmeter, and classifies stool according to the Bristol stool form scale using deep learning, with performance that is comparable to the performance of trained medical personnel. Each user of the toilet is identified through their fingerprint and the distinctive features of their anoderm, and the data are securely stored and analysed in an encrypted cloud server. The toilet may find uses in the screening, diagnosis and longitudinal monitoring of specific patient populations.
View details for DOI 10.1038/s41551-020-0534-9
View details for PubMedID 32251391
Microvesicle-mediated delivery of minicircle DNA results in effective gene-directed enzyme prodrug cancer therapy.
Molecular cancer therapeutics
An emerging approach for cancer treatment employs the use of extracellular vesicles (EVs), specifically exosomes and microvesicles, as delivery vehicles. We previously demonstrated that microvesicles can functionally deliver plasmid DNA to cells and showed that plasmid size and sequence, in part, determine the delivery efficiency. In this study, delivery vehicles comprised of microvesicles loaded with engineered minicircle (MC) DNA that encodes prodrug converting enzymes were developed as a cancer therapy in mammary carcinoma models. We demonstrated that MCs can be loaded into shed microvesicles with greater efficiency than their parental plasmid counterparts and that microvesicle-mediated MC delivery led to significantly higher and more prolonged transgene expression in recipient cells than microvesicles loaded with the parental plasmid. Microvesicles loaded with MCs encoding a thymidine kinase (TK)/nitroreductase (NTR) fusion protein produced prolonged TK-NTR expression in mammary carcinoma cells. In vivo delivery of TK-NTR and administration of prodrugs led to the effective killing of both targeted cells and surrounding tumor cells via TK-NTR-mediated conversion of co-delivered prodrugs into active cytotoxic agents. In vivo evaluation of the bystander effect in mouse models demonstrated that for effective therapy, at least 1% of tumor cells need to be delivered with TK-NTR-encoding MCs. These results suggest that MC delivery via microvesicles can mediate gene transfer to an extent that enables effective prodrug conversion and tumor cell death such that it comprises a promising approach to cancer therapy.
View details for DOI 10.1158/1535-7163.MCT-19-0299
View details for PubMedID 31451563
Comparison of precision and speed in laparoscopic and robot-assisted surgical task performance
JOURNAL OF SURGICAL RESEARCH
2018; 223: 29–33
Robotic platforms have the potential advantage of providing additional dexterity and precision to surgeons while performing complex laparoscopic tasks, especially for those in training. Few quantitative evaluations of surgical task performance comparing laparoscopic and robotic platforms among surgeons of varying experience levels have been done. We compared measures of quality and efficiency of Fundamentals of Laparoscopic Surgery task performance on these platforms in novices and experienced laparoscopic and robotic surgeons.Fourteen novices, 12 expert laparoscopic surgeons (>100 laparoscopic procedures performed, no robotics experience), and five expert robotic surgeons (>25 robotic procedures performed) performed three Fundamentals of Laparoscopic Surgery tasks on both laparoscopic and robotic platforms: peg transfer (PT), pattern cutting (PC), and intracorporeal suturing. All tasks were repeated three times by each subject on each platform in a randomized order. Mean completion times and mean errors per trial (EPT) were calculated for each task on both platforms. Results were compared using Student's t-test (P < 0.05 considered statistically significant).Among novices, greater errors were noted during laparoscopic PC (Lap 2.21 versus Robot 0.88 EPT, P < 0.001). Among expert laparoscopists, greater errors were noted during laparoscopic PT compared with robotic (PT: Lap 0.14 versus Robot 0.00 EPT, P = 0.04). Among expert robotic surgeons, greater errors were noted during laparoscopic PC compared with robotic (Lap 0.80 versus Robot 0.13 EPT, P = 0.02). Among expert laparoscopists, task performance was slower on the robotic platform compared with laparoscopy. In comparisons of expert laparoscopists performing tasks on the laparoscopic platform and expert robotic surgeons performing tasks on the robotic platform, expert robotic surgeons demonstrated fewer errors during the PC task (P = 0.009).Robotic assistance provided a reduction in errors at all experience levels for some laparoscopic tasks, but no benefit in the speed of task performance. Robotic assistance may provide some benefit in precision of surgical task performance.
View details for DOI 10.1016/j.jss.2017.07.037
View details for Web of Science ID 000425850400005
View details for PubMedID 29433882
- Toward achieving precision health SCIENCE TRANSLATIONAL MEDICINE 2018; 10 (430)
An intravascular magnetic wire for the high-throughput retrieval of circulating tumour cells in vivo.
Nature biomedical engineering
2018; 2: 696–705
The detection and analysis of rare blood biomarkers is necessary for early cancer diagnosis and to facilitate the development of tailored therapies. However, current methods for the isolation of circulating tumor cells (CTCs) or nucleic acids present in a standard clinical sample of only 5-10 mL of blood provide inadequate yields for early cancer detection and comprehensive molecular profiling. We have developed a flexible magnetic wire that can retrieve rare biomarkers from the subject's blood in vivo at a much higher yield. The wire is inserted and removed through a standard intravenous catheter and captures biomarkers that have been previously labeled with injected magnetic particles. In a proof-of-concept experiment in a live porcine model, we demonstrate the in vivo labeling and single-pass capture of viable model CTCs in less than 10 seconds. The wire achieves capture efficiencies that correspond to enrichments of 10-80 times the amount of CTCs in a 5-mL blood draw, and to 500-5,000 times the enrichments achieved by the commercially available Gilupi CellCollector.
View details for PubMedID 30524876
Sleep Disturbance and Fatigue Are Associated With More Severe Urinary Incontinence and Overactive Bladder Symptoms
2017; 109: 67–73
To investigate the relationship between sleep disturbance, fatigue, and urinary incontinence (UI) and overactive bladder (OAB) symptoms among patients with OAB.Patients who were diagnosed with OAB and age-matched control subjects without OAB were enrolled. Sleep disturbance and fatigue symptoms were assessed using the Patient-Reported Outcomes Measurement Information System (PROMIS) short forms. UI and OAB symptoms were assessed using the International Consultation on Incontinence Questionnaire-Urinary Incontinence (ICIQ-UI), the International Consultation on Incontinence Questionnaire-Overactive Bladder (ICIQ-OAB), the Overactive Bladder Questionnaire (OAB-q), the Urogenital Distress Inventory Short Form (UDI-6), and the Incontinence Impact Questionnaire Short Form (IIQ-7). Psychosocial health (depression, anxiety, and perceived stress level) was also assessed.Patients with OAB reported a significantly greater sleep disturbance compared with controls (PROMIS 8b T-scores: 54.3 ± 10.3 vs 43.8 ± 9.2). Patients with OAB also reported a significantly greater fatigue compared with controls (PROMIS 7a T-scores: 54.7 ± 9.6 vs 46.0 ± 6.4). After adjusting for nocturia, the differences in sleep disturbance between OAB and controls became insignificant (P = .21), whereas the differences in fatigue between OAB and controls remained significant (P = .014). Among patients with OAB, there were positive correlations between sleep disturbance and the severity of OAB symptoms (ICIQ-OAB), poorer health-related quality of life (OAB-q QOL), the severity of UI symptoms (ICIQ-UI), greater incontinence impact (IIQ-7), and urinary bother (UDI-6). Positive correlations were also observed between fatigue and worse UI and OAB symptoms and quality of life. Both sleep disturbance and fatigue were associated with poor psychosocial health (depression, anxiety, and higher stress level) among patients with OAB.Sleep disturbance and fatigue are present in substantial percentages of patients with OAB. Among patients with OAB, sleep disturbance and fatigue were associated with more severe UI and OAB symptoms, worse health-related quality of life, and poorer psychosocial health.
View details for DOI 10.1016/j.urology.2017.07.039
View details for Web of Science ID 000415597100014
View details for PubMedID 28826875
View details for PubMedCentralID PMC5669629
The Exosome Total Isolation Chip.
Circulating tumor-derived extracellular vesicles (EVs) have emerged as a promising source for identifying cancer biomarkers for early cancer detection. However, the clinical utility of EVs has thus far been limited by the fact that most EV isolation methods are tedious, nonstandardized, and require bulky instrumentation such as ultracentrifugation (UC). Here, we report a size-based EV isolation tool called ExoTIC (exosome total isolation chip), which is simple, easy-to-use, modular, and facilitates high-yield and high-purity EV isolation from biofluids. ExoTIC achieves an EV yield ∼4-1000-fold higher than that with UC, and EV-derived protein and microRNA levels are well-correlated between the two methods. Moreover, we demonstrate that ExoTIC is a modular platform that can sort a heterogeneous population of cancer cell line EVs based on size. Further, we utilize ExoTIC to isolate EVs from cancer patient clinical samples, including plasma, urine, and lavage, demonstrating the device's broad applicability to cancers and other diseases. Finally, the ability of ExoTIC to efficiently isolate EVs from small sample volumes opens up avenues for preclinical studies in small animal tumor models and for point-of-care EV-based clinical testing from fingerprick quantities (10-100 μL) of blood.
View details for DOI 10.1021/acsnano.7b04878
View details for PubMedID 29090896
Transfer and priming of surgical skills across minimally invasive surgical platforms
JOURNAL OF SURGICAL RESEARCH
2016; 206 (1): 48–52
Robot-assisted laparoscopic surgery (RALS) uses 3-dimensional visualization and wristed instruments that provide more degrees of freedom than rigid traditional laparoscopic (TLS) instrumentation. These features have been touted to improve accuracy and efficiency during surgical task performance. Little is known, however, about the transferability of skills between the two platforms or whether task performance on one platform primes surgeons for task performance on the other.Twenty-six subjects naïve to RALS were recruited to perform three Fundamentals of Laparoscopic Surgery tasks on both TLS and RALS platforms: peg transfer, pattern cutting (PC), and intracorporeal suturing. All tasks were performed within Fundamentals of Laparoscopic Surgery testing parameters and repeated three times by each subject on each platform. Platform and task order were randomized. Errors in task performance were defined as drops in the peg transfer task, faults 5 mm or more from the defined pattern during PC, and faults greater than 1 mm in suture placement from the defined points in intracorporeal suturing. Mean completion times and mean errors per trial (EPT) were calculated for each task on both platforms. Results were compared between those who performed TLS first (LF) and those who performed RALS first (RF) using unpaired Student's t-test (P < 0.05 considered statistically significant).No statistically significant differences in task completion time were noted between the LF and RF groups. RF subjects had fewer errors during robotic PC than LF subjects (1.02 EPT versus 1.86 EPT, respectively; P = 0.02). No other differences in task quality were noted.In surgeon's naïve to RALS, there is no evidence that skills acquired on RALS or TLS platforms are transferable to the other platform or that performing tasks on one platform primes a subject for task performance on the other. Performing TLS PC may have had a negative impact on subsequent RALS PC performance. These findings suggest that distinct programs for skills acquisition are necessary for both the TLS and RALS platforms.
View details for DOI 10.1016/j.jss.2016.06.026
View details for Web of Science ID 000387982900009
View details for PubMedID 27916374