JKJayesh Kamath

Jayesh Kamath

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Jayesh Kamath has a total of 54 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Predicting Depression Treatment Outcome Using Daily Step Count Sensory Data

    by Md Zakir Hossain, Soumyashree Sahoo, Chinmaey Shende, Parit Patel, Reynaldo Morillo, Zijie Pan, Xinyu Wang, Jinbo Bi, Jayesh Kamath, Alexander Russell, et al.

    Personalized depression treatment, despite decades of research, is still elusive. The current best practice is closely monitoring patients and adjusting the treatment as needed. Close monitoring of patients is, however, challenging in clinical settings due to high cost, limited resources, and patient burden. In this article, we explore using daily step counts automatically collected by a fitness wristband for depression treatment status monitoring. We focus on deep learning models that can effectively learn feature representation from a sequence of raw daily step count data, without relying on handcrafted features. One significant challenge is that our dataset is small, while deep learning models typically need large amounts of data to achieve strong generalization. To address this challenge, we investigate self-supervised contrastive pretraining to learn feature representation, followed by utilizing these representations for binary classification. Our results show that this contrastive pretraining approach significantly outperforms several state-of-the-art deep learning models. When using step count data together with a one-shot self-reported baseline questionnaire score, it leads to validation \(F_{1}\) score up to 0.74, comparable to the best \(F_{1}\) score (0.72) when using baseline and burdensome periodic self-reported scores. Last, when further including daily sleep data, the best \(F_{1}\) score is improved to 0.77.


  • Open Access

    Cross-platform Prediction of Depression Treatment Outcome Using Location Sensory Data on Smartphones

    by Soumyashree Sahoo, Zakir Hossain, Chinmaey Shende, Parit Patel, Yushuo Niu, Reynaldo Morillo, Xinyu Wang, Shweta Ware, Jinbo Bi, Jayesh Kamath, et al.

    Currently, depression treatment relies on closely monitoring patients’ response to treatment and adjusting the treatment as needed. Using self-reported or physician-administrated questionnaires to monitor treatment response is, however, subjective, costly and suffers from recall bias. In this paper, we explore using location sensory data collected passively on smartphones to predict treatment outcome. To address heterogeneous data collection on Android and iOS phones, the two predominant smartphone platforms, we explore using domain adaptation techniques to map their data to a common feature space, and then use the data jointly to train machine learning models. We further explore integrating contrastive learning with domain adaptation to augment data and learn feature embeddings. These learned embeddings are then used to train machine learning models to predict depression treatment outcomes. Our evaluation shows that using the embeddings learned by jointly integrating contrastive learning and domain adaptation leads to the best prediction accuracy. In addition, our results show that using location features and baseline self-reported questionnaire score can lead to F1 score up to 0.76. This accuracy is comparable to that obtained using periodic self-reported questionnaires, indicating that using location data is a promising direction for predicting depression treatment outcome. Last, when all location and questionnaire data are used together, the F1 score further increases to 0.79.


  • Open Access

    Smartphone Data Gathered Early in Depression Treatment Predicts Treatment Outcome

    by Soumyashree Sahoo, Zakir Hossain, Chinmaey Shende, Parit Patel, Xinyu Wang, Jinbo Bi, Jayesh Kamath, Alexander Russell, Dongjin Song, Bing Wang


  • Open Access

    Cross-platform Prediction of Depression Treatment Outcome Using Location Sensory Data on Smartphones

    by Soumyashree Sahoo, Chinmaey Shende, Zakir Hossain, Parit Patel, Yushuo Niu, Xinyu Wang, Shweta Ware, Jinbo Bi, Jayesh Kamath, Alexander Russel, et al.

    Currently, depression treatment relies on closely monitoring patients response to treatment and adjusting the treatment as needed. Using self-reported or physician-administrated questionnaires to monitor treatment response is, however, burdensome, costly and suffers from recall bias. In this paper, we explore using location sensory data collected passively on smartphones to predict treatment outcome. To address heterogeneous data collection on Android and iOS phones, the two predominant smartphone platforms, we explore using domain adaptation techniques to map their data to a common feature space, and then use the data jointly to train machine learning models. Our results show that this domain adaptation approach can lead to significantly better prediction than that with no domain adaptation. In addition, our results show that using location features and baseline self-reported questionnaire score can lead to F1 score up to 0.67, comparable to that obtained using periodic self-reported questionnaires, indicating that using location data is a promising direction for predicting depression treatment outcome.

  • Open Access

    Predicting Symptom Improvement During Depression Treatment Using Sleep Sensory Data

    by Chinmaey Shende, Soumyashree Sahoo, Stephen Sam, Parit Patel, Reynaldo Morillo, Xinyu Wang, Shweta Ware, Jinbo Bi, Jayesh Kamath, Alexander Russell, et al.

    Depression is a serious mental illness. The current best guideline in depression treatment is closely monitoring patients and adjusting treatment as needed. Close monitoring of patients through physician-administered follow-ups or self-administered questionnaires, however, is difficult in clinical settings due to high cost, lack of trained professionals, and burden to the patients. Sensory data collected from mobile devices has been shown to provide a promising direction for long-term monitoring of depression symptoms. Most existing studies in this direction, however, focus on depression detection; the few studies that are on predicting changes in depression are not in clinical settings. In this paper, we investigate using one type of sensory data, sleep data, collected from wearables to predict improvement of depression symptoms over time after a patient initiates a new pharmacological treatment. We apply sleep trend filtering to noisy sleep sensory data to extract high-level sleep characteristics and develop a family of machine learning models that use simple sleep features (mean and variation of sleep duration) to predict symptom improvement. Our results show that using such simple sleep features can already lead to validation F1 score up to 0.68, indicating that using sensory data for predicting depression improvement during treatment is a promising direction.


  • Open Access

    The Prediction of Suicidal Ideation as a Function of Daily Mood and Anxiety Scores Collected Using mHealth Technology in Patients Undergoing Treatment for Depression

    by Parit Patel, Grace Chan, Bing Wang, Jayesh Kamath

    Background: Suicide is one of the most common causes of death in the United States. The rates of suicide have increased by 33% in the period 1999-2019. In 2019, suicide was responsible for one death every 11 minutes. Clinically, a previous history of suicide attempts is the main risk factor, as well as the presence of comorbid conditions like depression and anxiety. Accurate and real-time prediction of suicidal thoughts may lead to improved management of patients with depression. Prediction of suicidality is difficult due to its day-to-day variability in relation to mood and anxiety symptoms. This can be overcome with the advent of mobile health (mHealth) technology that can capture granular data scores at a higher frequency than conventional therapeutic visitations. Objective: The aim of this study is to predict suicidal ideation using self-reported mood and anxiety in patients undergoing treatment for depression. Methods: This study will use data from the DepWatch study, an mHealth study that uses the DepWatch app developed by our research group. The objective of this longitudinal study is to develop an mHealth-based, personalized diagnostic prediction system for patients undergoing treatment for depression. Patients are followed over 12 weeks using electronic assessments conducted via the DepWatch app installed on their smartphones. The electronic assessments include the Quick Inventory of Depression Symptomatology-Self Report (QIDS-SR), conducted on a weekly basis, and weekly medication adherence and medication safety and tolerability questionnaires. The assessments include brief mood and anxiety assessments conducted on a daily basis. Generalized estimating equation modeling for a binary outcome (the presence or absence of suicidal thought), clustered by individual subjects, will be used. The key explanatory variables are the daily mood and anxiety levels (time variant). The outcome variable is suicidal ideation as determined by the self-reported subject response to question 12 (about suicidality) on the QIDS-SR scale. Variables that show significance at P values <.1 are subsequently used in the multivariate model, in addition to mood and anxiety. Results: A total of 34 subjects are in the interim analysis set. The median age is 26 years, 85.3% are female, and 64.7% are White. In addition, 38.2% either have a college degree or graduate education, while 23.5% are unemployed and 41.1% are full-time employed. About half of participants (52.9%) earn less than $50,000 annually and 61% have never smoked. Univariate analysis shows statistical significance of gender, race, aggregate anxiety, and employment status at .1 significance level. In the multivariate model, only gender and employment status are significant at .05. Race is marginally insignificant (P=.068). Conclusions: Suicidality and completed suicides are significant public health problems, especially in patients with depression. The mHealth technology and statistical modelling that captures daily variability in anxiety leading up to suicidal ideation can help predict suicidality.


  • Open Access

    Digital phenotyping in depression diagnostics: Integrating psychiatric and engineering perspectives

    by Jayesh Kamath, Roberto Leon Barriera, Neha Jain, Efraim Keisari, Bing Wang

    Digital phenotyping in depression diagnostics: Integrating psychiatric and engineering perspectives


  • Open Access

    Grant Report on SCH: Personalized Depression Treatment Supported by Mobile Sensor Analytics

    by Jayesh Kamath, Jinbo Bi, Alexander Russell, Bing Wang

    We report on the newly started project “SCH: Personalized Depression Treatment Supported by Mobile Sensor Analytics”. The current best practice guidelines for treating depression call for close monitoring of patients, and periodically adjusting treatment as needed. This project will advance personalized depression treatment by developing a system, DepWatch, that leverages mobile health technologies and machine learning tools. The objective of DepWatch is to assist clinicians with their decision making process in the management of depression. The project comprises two studies. Phase I collects sensory data and other data, e.g., clinical data, ecological momentary assessments (EMA), tolerability and safety data from 250 adult participants with unstable depression symptomatology initiating depression treatment. The data thus collected will be used to develop and validate assessment and prediction models, which will be incorporated into DepWatch system. In Phase II, three clinicians will use DepWatch to support their clinical decision making process. A total of 128 participants under treatment by the three participating clinicians will be recruited for the study. A number of new machine learning techniques will be developed.


    • 1 June 2016
      • Vol. 44  (2) , 158-63

    Atomoxetine for the Treatment of ADHD in Incarcerated Adolescents.

    by Sarah Jillani, Prina Patel, Robert Trestman, Jayesh Kamath

    Effective interventions for adolescents with attention deficit/hyperactivity disorder (ADHD) in the correctional setting may improve care during incarceration, decrease risk of substance relapse, and reduce recidivism after release from the correctional setting of these individuals. The present report delineates the epidemiology of adolescent ADHD in the correctional setting and its association with substance use disorders and comorbid psychiatric illnesses. Evidence suggests that adolescents with ADHD have a higher risk of arrest and incarceration during adulthood. The present report examines evidence related to efficacy of atomoxetine, a nonstimulant medication for the treatment of adolescent ADHD, and presents data from a case series evaluating the effectiveness of atomoxetine for the treatment of adolescent ADHD in the Connecticut correctional setting. The results from the case series suggest that atomoxetine is effective for the treatment of adolescent ADHD in the context of significant past substance use. In summary, adolescents with ADHD have an elevated risk of incarceration and developing substance use disorders. The present review and pilot case series suggest that atomoxetine is an effective treatment for adolescents with ADHD in the correctional setting.