Jin Zhou

Jin Zhou

Auto-generated profile, based on publicly available information

Jin Zhou

Duke University

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
12 June 2020 in IEEE Transactions on Biomedical Circuits and Systems
Verified
Journal Article
Open access
JZJin Zhou
Jin Zhou
CWClaire M. Welling
Claire M. Welling
MVMariana M. Vasquez
Mariana M. Vasquez
SGSonia Grego
Sonia Grego
KCKrishnendu Chakrabarty
Krishnendu Chakrabarty
There is an unmet need for a low-cost instrumented technology for detecting sanitation-related malodor as an alert for maintenance around shared toilets and emerging technologies for onsite waste treatment. In this article, our approach to an electronic nose for sanitation-related malodor is based on the use of electrochemical gas sensors, and machine-learning techniques for sensor selection and odor classification. We screened 10 sensors from different vendors with specific target gases and recorded their response to malodor from fecal specimens and urine specimens, and confounding good odors such as popcorn. The analysis of 180 odor exposures data by two feature-selection methods based on mutual information indicates that, for malodor detection, the decision tree (DT) classifier with seven features from four sensors provides 88.0% balanced accuracy and 85.1% macro F1 score. However, the k-nearest-neighbors (KNN) classifier with only three features (from two sensors) obtains 83.3% balanced accuracy and 81.3% macro F1 score. For classification of urine against feces malodor, a balanced accuracy of 94.0% and a macro F1 score of 92.9% are achieved using only four features from three sensors and a logistic regression (LR) classifier.

11 March 2022 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
JZJin Zhou
Jin Zhou
JMJackson McNabb
Jackson McNabb
NDNick DeCapite
Nick DeCapite
JRJose R. Ruiz
Jose R. Ruiz
DFDeborah A. Fisher
Deborah A. Fisher
KCKrishnendu Chakrabarty
Krishnendu Chakrabarty
Gastrointestinal (GI) conditions are widespread and significantly impact the quality of life and healthcare. Stool appearance is a valuable GI diagnostic indicator, and with the advent of the Internet of Things (IoT), daily monitoring of excreta from a toilet is emerging as a promising digital health tool. This article describes a stool image analysis approach that classifies two physiologically relevant stool characteristics: 1) form and 2) color for an IoT-based smart toilet. We constructed a stool image data set with 3275 images, spanning all seven types of the Bristol stool-form scale (BSFS), a widely used metric for stool consistency and a variety of colors. We used ground-truth data obtained through the annotation of our data set by two gastroenterologists and developed a stool-color card to standardize the labeling of stool colors. We addressed two limitations associated with the application of computer-vision techniques to a smart-toilet system: 1) uneven separability between different stool-form categories and 2) class imbalance in the data set. We present results on hierarchical convolutional neural network (CNN) architectures for training a stool-form classifier and on perceptual color quantization coupled with machine-learning techniques to optimize the color-feature space for the classification of stool color. We utilized an edge–cloud approach to pursue an optimal balance between accuracy and latency and for the classification of stool form, we achieved a balanced accuracy of 84.4% and 84.2% reduction in latency compared to a cloud-model only. For color classification, the logistic-regression (LR) classifier provided 80.8% balanced accuracy and 47.3% reduction in communication latency.

1 September 2024 in Journal of Medical Imaging
Verified
Journal Article
Open access
PDF file
JZJin Zhou
Jin Zhou
XLXiang Li
Xiang Li
DDDawit Demeke
Dawit Demeke
TDTimothy A. A. Dinh
Timothy A. A. Dinh
YYYingbao Yang
Yingbao Yang
KLKyle J. J. Lafata
Kyle J. J. Lafata
Previous studies have shown that the severity of arterio- and arteriolosclerosis is associated with the risk of kidney disease progression across various disease etiologies.1 – 3 Unlike atherosclerosis, which is a specific form of arterial thickening or hardening caused by atheroma plaque build-up in the inner lining of large arteries such as the carotid artery, arterio- and arteriolosclerosis affects medium- and small-sized arteries and arterioles within the kidney parenchyma. Arterio- and arteriolosclerosis, characterized by fibrous thickening of the intima and hypertrophy of the media, and arteriolar hyalinosis, a lesion of eosin and periodic acid–schiff (PAS)-positive amorphous material deposited in the vessel wall, are manifestations of chronic damage of kidney arteries and arterioles.4,5 Thus, the quantification of chronic vascular changes present in renal biopsies is relevant to the diagnosis and prognostication of kidney failure and to informing treatment selection.5 – 7 In clinical practice, biopsies are visually inspected to semi-quantitatively assess the extent of intima and media thickening.5 However, this approach may be influenced by inherent cognitive and visual biases,8 with the precision and accuracy of these observations limited by high inter- and intra-observer variability.9 – 12

8 October 2025 in Journal of the American Society of Nephrology
Verified
Journal Article
Open access
XLXiang Li
Xiang Li
MSManav Shah
Manav Shah
QLQian Liu
Qian Liu
JZJin Zhou
Jin Zhou
GSGina Sotolongo
Gina Sotolongo
LBLaura Barisoni
Laura Barisoni
Key Points: Graph topology features of inflammation can enhance prognostication of proteinuric glomerular diseases. Computational image analysis is useful for tissue interrogation and extraction of hidden tissue characteristics. Digital pathology and computer vision allowed for characterization of inflammation patterns beyond human vision. Background: The distribution of inflammation in the kidney and its clinical relevance is understudied. This study aimed to computationally quantify lymphocyte topology and test its prediction of disease progression. Methods: NEPhrotic syndrome sTUdy NEtwork (NEPTUNE) ( N =333)/Cure Glomerulonephropathy (CureGN) participants ( N =155 focal segmental glomerulosclerosis, N =178 minimal change disease) with available clinical/demographic data and one hematoxylin and eosin–stained whole slide image were included. Cortex and lymphocytes were automatically segmented. A novel graph-based clustering algorithm was applied to identify dense versus sparse lymphocytic habitats, from which 26 pathomic features were extracted to capture cell density, connectivity, clustering, and centrality. The association of these pathomic features with disease progression (40% eGFR decline or KRT) was assessed using ElasticNet-regularized Cox proportional hazards models. Clinical and demographic characteristics and percent of interstitial fibrosis and inflammation were added as potential confounders. Kaplan–Meier survival analysis with log-rank test was used to evaluate risk stratification. Two validation strategies were applied: ( 1 ) training on NEPTUNE with external validation on CureGN data and ( 2 ) using an internal bootstrap validation of the combined datasets for training and validation, respectively. Results: Unadjusted analysis: 17 features (65%) retained significance after adjustment for standard clinicodemographic variables, Number of K-core in sparse habitat maintained significance (hazard ratio, 1.29; 95% confidence interval, 1.04 to 1.61) even after adjustment for lymphocyte density, and Average Degree in dense habitat had borderline significance (hazard ratio, 1.25; 95% confidence interval, 1.00 to 1.57) after adjustment for interstitial fibrosis. Multivariable models (clinical/demographic+graph features) achieved validation concordance index of 0.78±0.15 in the CureGN external validation and 0.77±0.06 in the combined internal validation dataset. Time-dependent discrimination showed consistent performance at 3-year (area under the time-varying receiver operating characteristic curve: 0.78 versus 0.76) and 5-year time points (area under the time-varying receiver operating characteristic curve: 0.74 versus 0.76) across validation strategies. Sparse habitat clustering patterns ( Maximum of K-core×Number of K-core in sparse habitat : 88% selection frequency) and dense habitat connectivity ( Average Degree in dense habitat: 47% selection frequency) were consistently identified as robust predictors alongside clinical variables. Conclusions: The topologic characterization of lymphocytic inflammation identified immune habitats, capturing the complexity of patterns of inflammation.

1 November 2024 in Neurogastroenterology & Motility
Verified
Journal Article
Closed access
JZJin Zhou
Jin Zhou
YLYuying Luo
Yuying Luo
JDJulia W. Darcy
Julia W. Darcy
KLKyle J. Lafata
Kyle J. Lafata
JRJose R. Ruiz
Jose R. Ruiz
SGSonia Grego
Sonia Grego

26 December 2025
Verified
Book Chapter
Closed access
PDF file
JZJin Zhou
Jin Zhou
JMJackson McNabb
Jackson McNabb
NDNick DeCapite
Nick DeCapite
JRJose R. Ruiz
Jose R. Ruiz
DFDeborah A. Fisher
Deborah A. Fisher
KCKrishnendu Chakrabarty
Krishnendu Chakrabarty