Tushar Mungle is postdoctoral scholar in Boussard lab at Stanford Center for Biomedical Informatics Research. He received bachelor's and master's degree in computer science followed by Ph.D. in clnical informatics. His work involve deducing inferences from electronic health records/clinical data and provide feasible solutions to existing bedside or physician encountered clinical problems. Previously at Tata Medical Center Kolkata, India, he was involved in analyzing pediatric acute lymphoblastic leukemia maintenance therapy (MT) data and developed tools to standardize the MT practice. Additionally, he investigated clinical data pertaining to ICiCLe-ALL-14 clinical trial sub-studies, and gallbladder cancer. He is extensively trained in R and python programming languages, statistics, longitudinal data analysis and predictive modeling.

Honors & Awards

  • Special Award for Oral Presentation, XIIIth SIOP Asia Conference (2021)
  • Qualified "Graduate Aptitude Test in Engineering" (GATE), Department of Higher Education, Ministry of Education (MoE), Government of India. (2012)
  • University Rank 3rd, Bachelor of Engineering, RTMN University (2011)

Professional Education

  • Ph.D., Indian Institute of Technology Kharagpur, Clinical Informatics (2020)
  • Master of Technology, Manipal Institute of Technology, Computer Science and Technology (2014)
  • Bachelor of Engineering, SVPCET, RTMN University, Computer Engineering (2011)

Stanford Advisors

Research Interests

  • Data Sciences

Current Research and Scholarly Interests

Use electronic health records (EHRs) to identify and classify common ocular diseases such as glaucoma, diabetic retinopathy, and macular degeneration. We aim to develop an approach to accurately identify these conditions using EHRs. This will be followed by cluster analysis to identify novel subtypes of these conditions that have not been recognized before. Finally, we will develop an approach to extract outcome data from EHRs for patients with these conditions in the primary care setting.

All Publications

  • Maintenance Treatment in Acute Lymphoblastic Leukemia: A Clinical Primer. Indian journal of pediatrics Krishnan, S., Mahadevan, A., Mungle, T., Gogoi, M. P., Saha, V. 2023


    Cure rates in pediatric acute lymphoblastic leukemia (ALL) currently approach 90% in the developed world. Treatment involves 6-8 mo of intensive multi-drug chemotherapy followed by 24 mo of maintenance treatment (ALL-MT). The cornerstone of ALL-MT is the daily administration of oral 6-mercaptopurine (6MP), a purine analogue. 6MP is combined with weekly oral methotrexate (MTX), an antifolate drug, to augment therapeutic activity. Some protocols include additional chemotherapy drugs (such as vincristine and corticosteroids) during MT. The objective of ALL-MT is to ensure uninterrupted treatment at the highest tolerated doses of 6MP and MTX. This requires periodic adjustments of 6MP and MTX doses throughout treatment. Tolerance is determined through regular clinical assessments and careful monitoring of blood counts. Tolerated drug doses vary widely among patients, influenced by genetic and non-genetic factors, and require individualized dosing. Suboptimal treatment intensity in ALL-MT is associated with inferior outcomes and results from failure to treat at highest tolerated drug doses and/or interruptions in treatment due to non-adherence or toxicity. Management of MT thus requires close supervision to ensure treatment adherence, periodic drug dose modifications, and treatment to tolerance, while minimizing treatment interruptions due to toxicity. The review highlights these challenges and discusses approaches and strategies for the management of MT, focusing on the Indian context.

    View details for DOI 10.1007/s12098-023-04687-6

    View details for PubMedID 37493925

  • Characteristics and outcomes of gallbladder cancer patients at the Tata Medical Center, Kolkata 2017-2019 CANCER MEDICINE Dutta, A., Mungle, T., Chowdhury, N., Banerjee, P., Gehani, A., Sen, S., Mallath, M., Roy, P., Krishnan, S., Ganguly, S., Banerjee, S., Roy, M., Saha, V. 2023: 9293-9302


    The north and north-eastern regions of India have among the highest incidence of gallbladder cancer (GBC) in the world. We report the clinicopathological charateristics and outcome of GBC patients in India.Electronic medical records of patients diagnosed with GBC at Tata Medical Center, Kolkata between 2017 and 2019 were analyzed.There were 698 cases of confirmed GBC with a median age of 58 (IQR: 50-65) years and female:male ratio of 1.96. At presentation, 91% (496/544) had stage III/IV disease and 30% (189/640) had incidental GBC. The 2-year overall survival (OS) was 100% (95% CI: 100-100); 61% (95% CI: 45-83); 30% (95% CI: 21-43); and 9% (95% CI: 6-13) for stages I-IV, respectively (p = <0.0001).   For all patients, the 2-year OS in patients who had a radical cholecystectomy followed by adjuvant therapy (N = 36) was 50% (95% CI: 39-64), compared to 29% (95% CI: 22-38) for those who had a simple cholecystectomy and/or chemotherapy (N = 265) and 9% (95% CI: 6-14) in patients who were palliated (N = 107) (p = <0.0001).The combined surgical/chemotherapy approach for patients with stage II GBC showed the best outcomes. Early detection of GBC remains problematic with the majority of patients presenting with stage III-IV and who have a median survival of 9.1 months. Our data suggests that the tumor is chemoresponsive and multi-center collaborative clinical trials to identify alternative therapies are urgently required.

    View details for DOI 10.1002/cam4.5677

    View details for Web of Science ID 000929911000001

    View details for PubMedID 36779618

    View details for PubMedCentralID PMC10166897

  • Comparative treatment costs of risk-stratified therapy for childhood acute lymphoblastic leukemia in India CANCER MEDICINE Mungle, T., Das, N., Pal, S., Gogoi, M., Das, P., Ghara, N., Ghosh, D., Arora, R., Bhakta, N., Saha, V., Krishnan, S. 2022: 3499-3508


    To evaluate the treatment cost and cost effectiveness of a risk-stratified therapy to treat pediatric acute lymphoblastic leukemia (ALL) in India.The cost of total treatment duration was calculated for a retrospective cohort of ALL children treated at a tertiary care facility. Children were risk stratified into standard (SR), intermediate (IR) and high (HR) for B-cell precursor ALL, and T-ALL. Cost of therapy was obtained from the hospital electronic billing systems and details of outpatient (OP) and inpatient (IP) from electronic medical records. Cost effectiveness was calculated in disability-adjusted life years.One hundred and forty five patients, SR (50), IR (36), HR (39), and T-ALL (20) were analyzed. Median cost of the entire treatment for SR, IR, HR, and T-ALL was found to be $3900, $5500, $7400, and $8700, respectively, with chemotherapy contributing to 25%-35% of total cost. Out-patient costs were significantly lower for SR (p < 0.0001). OP costs were higher than in-patient costs for SR and IR, while in-patient costs were higher in T-ALL. Costs for non-therapy admissions were significantly higher in HR and T-ALL (p < 0.0001), representing over 50% of costs of in-patient therapy. HR and T-ALL also had longer durations of non-therapy admissions. Based on WHO-CHOICE guidelines, the risk-stratified approach was very cost effective for all categories of patients.Risk-stratified approach to treat childhood ALL is very cost-effective for all categories in our setting. The cost for SR and IR patients is significantly reduced through decreased IP admissions for both, chemotherapy and non-chemotherapy reasons.

    View details for DOI 10.1002/cam4.5140

    View details for Web of Science ID 000851532700001

    View details for PubMedID 36812120

    View details for PubMedCentralID PMC9939102

  • Allopurinol adjuvant in acute lymphoblastic leukaemia maintenance treatment 25th Annual Conference of the Pediatric Hematology Oncology Kamle, A., Ghara, N., Ghosh, D., Gogoi, M., Jana, B., Mungle, T. 2022: S48
  • Developing an automated dose advice programme to assist adaptive antimetabolite dose decisions during maintenance therapy in acute lymphoblastic leukaemia XIIIth SIOP ASIA CONFERENCE Mungle, T., Gogoi, M., Mitra, S., Poddar, M., Roy, P., Bhattacharya, B., Mukhopadhyay, J., Saha, V., Bhattacharya, S., Krishnan, S. 2020: S10
  • Web-Enabled Distributed Health-Care Framework for Automated Malaria Parasite Classification: an E-Health Approach JOURNAL OF MEDICAL SYSTEMS Maity, M., Dhane, D., Mungle, T., Maiti, A. K., Chakraborty, C. 2017; 41 (12): 192


    Web-enabled e-healthcare system or computer assisted disease diagnosis has a potential to improve the quality and service of conventional healthcare delivery approach. The article describes the design and development of a web-based distributed healthcare management system for medical information and quantitative evaluation of microscopic images using machine learning approach for malaria. In the proposed study, all the health-care centres are connected in a distributed computer network. Each peripheral centre manages its' own health-care service independently and communicates with the central server for remote assistance. The proposed methodology for automated evaluation of parasites includes pre-processing of blood smear microscopic images followed by erythrocytes segmentation. To differentiate between different parasites; a total of 138 quantitative features characterising colour, morphology, and texture are extracted from segmented erythrocytes. An integrated pattern classification framework is designed where four feature selection methods viz. Correlation-based Feature Selection (CFS), Chi-square, Information Gain, and RELIEF are employed with three different classifiers i.e. Naive Bayes', C4.5, and Instance-Based Learning (IB1) individually. Optimal features subset with the best classifier is selected for achieving maximum diagnostic precision. It is seen that the proposed method achieved with 99.2% sensitivity and 99.6% specificity by combining CFS and C4.5 in comparison with other methods. Moreover, the web-based tool is entirely designed using open standards like Java for a web application, ImageJ for image processing, and WEKA for data mining considering its feasibility in rural places with minimal health care facilities.

    View details for DOI 10.1007/s10916-017-0834-0

    View details for Web of Science ID 000413673900008

    View details for PubMedID 29075939

  • Fuzzy spectral clustering for automated delineation of chronic wound region using digital images COMPUTERS IN BIOLOGY AND MEDICINE Dhane, D., Maity, M., Mungle, T., Bar, C., Achar, A., Kolekar, M., Chakraborty, C. 2017; 89: 551-560


    Chronic wound is an abnormal disease condition of localized injury to the skin and its underlying tissues having physiological impaired healing response. Assessment and management of such wound is a significant burden on the healthcare system. Currently, precise wound bed estimation depends on the clinical judgment and remains a difficult task. The paper introduces a novel method for ulcer boundary demarcation and estimation, using optical images captured by a hand-held digital camera. The proposed approach involves gray based fuzzy similarity measure using spatial knowledge of an image. The fuzzy measure is used to construct similarity matrix. The best color channel was chosen by calculating the mean contrast for 26 different color channels of 14 color spaces. It was found that Db color channel has highest mean contrast which provide best segmentation result in comparison with other color channels. The fuzzy spectral clustering (FSC) method was applied on Db color channel for effective delineation of wound region. The segmented wound regions were effectively post-processed using various morphological operations. The performance of proposed segmentation technique was validated by ground-truth images labeled by two experienced dermatologists and a surgeon. The FSC approach was tested on 70 images. FSC effectively segmented targeted ulcer boundary yielding 91.5% segmentation accuracy, 86.7%, Dice index and 79.0%. Jaccard score. The sensitivity and specificity was found to be 87.3% and 95.7% respectively. The performance evaluation shows the robustness of the proposed method of wound area segmentation and its potential to be used for designing patient comfort centric wound care system.

    View details for DOI 10.1016/j.compbiomed.2017.04.004

    View details for Web of Science ID 000413376600054

    View details for PubMedID 28479109

  • MRF-ANN: a machine learning approach for automated ER scoring of breast cancer immunohistochemical images JOURNAL OF MICROSCOPY Mungle, T., Tewary, S., Das, D. K., Arun, I., Basak, B., Agarwal, S., Ahmed, R., Chatterjee, S., Chakraborty, C. 2017; 267 (2): 117-129


    Molecular pathology, especially immunohistochemistry, plays an important role in evaluating hormone receptor status along with diagnosis of breast cancer. Time-consumption and inter-/intraobserver variability are major hindrances for evaluating the receptor score. In view of this, the paper proposes an automated Allred Scoring methodology for estrogen receptor (ER). White balancing is used to normalize the colour image taking into consideration colour variation during staining in different labs. Markov random field model with expectation-maximization optimization is employed to segment the ER cells. The proposed segmentation methodology is found to have F-measure 0.95. Artificial neural network is subsequently used to obtain intensity-based score for ER cells, from pixel colour intensity features. Simultaneously, proportion score - percentage of ER positive cells is computed via cell counting. The final ER score is computed by adding intensity and proportion scores - a standard Allred scoring system followed by pathologists. The classification accuracy for classification of cells by classifier in terms of F-measure is 0.9626. The problem of subjective interobserver ability is addressed by quantifying ER score from two expert pathologist and proposed methodology. The intraclass correlation achieved is greater than 0.90. The study has potential advantage of assisting pathologist in decision making over manual procedure and could evolve as a part of automated decision support system with other receptor scoring/analysis procedure.

    View details for DOI 10.1111/jmi.12552

    View details for Web of Science ID 000405746400001

    View details for PubMedID 28319275

  • An Ensemble Rule Learning Approach for Automated Morphological Classification of Erythrocytes JOURNAL OF MEDICAL SYSTEMS Maity, M., Mungle, T., Dhane, D., Maiti, A. K., Chakraborty, C. 2017; 41 (4): 56


    The analysis of pathophysiological change to erythrocytes is important for early diagnosis of anaemia. The manual assessment of pathology slides is time-consuming and complicated regarding various types of cell identification. This paper proposes an ensemble rule-based decision-making approach for morphological classification of erythrocytes. Firstly, the digital microscopic blood smear images are pre-processed for removal of spurious regions followed by colour normalisation and thresholding. The erythrocytes are segmented from background image using the watershed algorithm. The shape features are then extracted from the segmented image to detect shape abnormality present in microscopic blood smear images. The decision about the abnormality is taken using proposed multiple rule-based expert systems. The deciding factor is majority ensemble voting for abnormally shaped erythrocytes. Here, shape-based features are considered for nine different types of abnormal erythrocytes including normal erythrocytes. Further, the adaptive boosting algorithm is used to generate multiple decision tree models where each model tree generates an individual rule set. The supervised classification method is followed to generate rules using a C4.5 decision tree. The proposed ensemble approach is precise in detecting eight types of abnormal erythrocytes with an overall accuracy of 97.81% and weighted sensitivity of 97.33%, weighted specificity of 99.7%, and weighted precision of 98%. This approach shows the robustness of proposed strategy for erythrocytes classification into abnormal and normal class. The article also clarifies its latent quality to be incorporated in point of care technology solution targeting a rapid clinical assistance.

    View details for DOI 10.1007/s10916-017-0691-x

    View details for Web of Science ID 000399826300009

    View details for PubMedID 28247304

  • Automated characterization and counting of Ki-67 protein for breast cancer prognosis: A quantitative immunohistochemistry approach COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE Mungle, T., Tewary, S., Arun, I., Basak, B., Agarwal, S., Ahmed, R., Chatterjee, S., Maity, A., Chakraborty, C. 2017; 139: 149-161


    Ki-67 protein expression plays an important role in predicting the proliferative status of tumour cells and deciding the future course of therapy in breast cancer. Immunohistochemical (IHC) determination of Ki-67 score or labelling index, by estimating the fraction of Ki67 positively stained tumour cells, is the most widely practiced method to assess tumour proliferation (Dowsett et al. 2011). Accurate manual counting of these cells (specifically nuclei) due to complex and dense distribution of cells, therefore, becomes critical and presents a major challenge to pathologists. In this paper, we suggest a hybrid clustering algorithm to quantify the proliferative index of breast cancer cells based on automated counting of Ki-67 nuclei. The proposed methodology initially pre-processes the IHC images of Ki-67 stained slides of breast cancer. The RGB images are converted to grey, L*a*b*, HSI, YCbCr, YIQ and XYZ colour space. All the stained cells are then characterized by two stage segmentation process. Fuzzy C-means quantifies all the stained cells as one cluster. The blue channel of the first stage output is given as input to k-means algorithm, which provides separate cluster for Ki-67 positive and negative cells. The count of positive and negative nuclei is used to calculate the F-measure for each colour space. A comparative study of our work with the expert opinion is studied to evaluate the error rate. The positive and negative nuclei detection results for all colour spaces are compared with the ground truth for validation and F-measure is calculated. The F-measure for L*a*b* colour space (0.8847) provides the best statistical result as compared to grey, HSI, YCbCr, YIQ and XYZ colour space. Further, a study is carried out to count nuclei manually and automatically from the proposed algorithm with an average error rate of 6.84% which is significant. The study provides an automated count of positive and negative nuclei using L*a*b*colour space and hybrid segmentation technique. Computerized evaluation of proliferation index can aid pathologist in assessing breast cancer severity. The proposed methodology, further, has the potential advantage of saving time and assisting in decision making over the present manual procedure and could evolve as an assistive pathological decision support system.

    View details for DOI 10.1016/j.cmpb.2016.11.002

    View details for Web of Science ID 000395223200013

    View details for PubMedID 28187885

  • A Secure One-Time Password Authentication Scheme Using Image Texture Features Maity, M., Dhane, D., Mungle, T., Chakraborty, R., Deokamble, V., Chakraborty, C., Mueller, P., Thampi, S. M., Bhuiyan, M. Z., Ko, R., Doss, R., Calero, J. M. SPRINGER-VERLAG SINGAPORE PTE LTD. 2016: 283-294