SASoumyadipta Acharya

Soumyadipta Acharya

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Soumyadipta Acharya has a total of 147 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Teaching Global Health Innovation: An Integrated, Immersive, and Iterative Approach

    by Sofía García del Barrio Cervera, Sabrina Zhou, Kunal S. Parikh, Marina Rincon Torroella, Youseph Yazdi, Soumyadipta Acharya

  • Open Access

    Deep learning convolutional neural networks for morphological identification of field-caught Anopheles stephensi in Ethiopia, using smartphone images of specimens

    by Parthvi Mehta, Aravind Sunil Kumar, Atul Zacharias, Aryaman Shodhan, Deming Li, Sunny M. Patel, Marina Rincon Torroella, Joanne M. Cunningham, Diane Lovin, Biniam Lukas Chbute, et al.

  • Open Access

    An AI‑enabled, microscope‑integrated tool for decentralizing breast cancer diagnosis in Uganda: A mixed‑methods usability and workflow feasibility study (Preprint)

    by Sofía García del Barrio Cervera, Sierra Perez, Krishna Tejaswini Sathi, Fujia Zheng, Kim Hwang Yeo, Shreya Jindal, Ava Taylor, Kenneth Mark Nova, Arjun Menta, Kara Nghiem, et al.

    BACKGROUND Breast cancer is a leading cause of cancer morbidity and mortality worldwide, with disproportionately poor outcomes in low- and middle-income countries. In Uganda, delays between clinical presentation and diagnosis are driven by limited pathology capacity, long specimen transport times and weak referral systems. A task-shifting, low-cost diagnostic tool that brings cytology assessment closer to patients could reduce diagnostic delay and loss to follow-up. OBJECTIVE To evaluate and compare the usability and workflow feasibility of two Ekyaalo Diagnostics hardware form factors (V1 and V2), a low-cost AI-assisted platform for decentralized fine needle aspiration cytology slide digitization and remote pathology review across urban and rural health facilities in Uganda. It consists of a standard light microscope, off-the-shelf optical components and a smartphone application. METHODS We conducted a parallel, mixed-methods usability study to compare two hardware form factors across clinical settings in Uganda, in partnership with Makerere University, Mulago National Referral Hospital (NRH) and Mbarara Regional Referral Hospital (RRH). Study 1 evaluated V1 (smartphone adapter) at two tertiary hospitals (Mulago NRH and Mbarara RRH), while Study 2 evaluated V2 (eyepiece camera) across 11 rural and urban healthcare facilities spanning Health Centre IV–VII levels. Outcomes included System Usability Scale (SUS), ROI capture performance, Likert-scale usability measures and qualitative feedback on workflow integration and AI-assisted diagnosis. Non-parametric tests were used for group comparisons, and thematic analysis was used to analyze qualitative data from interviews and direct observations. RESULTS Both studies yielded encouraging results. In Study 1, mean SUS was 77.24 ± 17.18 indicated “good” usability and no statistically significant SUS differences across roles. Among 38 participants with capture data, mean regions of interest (ROI) captured was 6.39 ± 9.66, approximating routine review. AI outputs were viewed as supportive of the pathologist, rather than definitive. In Study 2, mean SUS was 66.59 ± 98.87, reflecting increased perceived complexity in rural contexts. Around 80% reported low confidence in acting on AI outputs without pathologist confirmation. Qualitative findings highlighted improved image quality and setup with V2 eyepiece camera but persistent constraints: shared microscopes, variable slide quality and a need for scanning the whole‑slide rather than sampling ROIs. CONCLUSIONS Ekyaalo Diagnostics demonstrated promising usability and strong acceptance among referral laboratory staff in Uganda. Ongoing iterative refinements that prioritize integration in rural settings, particularly by providing broader slide context and minimizing disruption to shared microscopes, are expected to enhance performance and acceptability. Given the similar challenges faced by low- and middle- income countries, Ekyaalo Diagnostics has the potential to reduce diagnostic delays in low-resource settings worldwide by supporting decentralized diagnostic workflows in low-resource settings.

  • Open Access

    Integration of VectorCam, an AI-Enabled Mosquito Surveillance Tool, into Uganda’s National Malaria Elimination Program: Protocol for a Mixed Methods Hybrid Type III Implementation Study (Preprint)

    by Mei-Li Hey, Marina Rincon-Torroella, Diwakar Mohan, Sunny Patel, Winnifred Kansiime, Aryaman Shodhan, Catherine Maiteki Sebuguzi, Youseph Yazdi, Peter Waiswa, Soumyadipta Acharya

    Background: Malaria remains one of the most significant global health threats, and its control depends on timely, accurate entomological surveillance. In Uganda, routine vector surveillance relies heavily on microscope-based mosquito identification by a limited number of trained entomologists, which can delay reporting. VectorCam is an AI-enabled digital tool to support real-time mosquito identification and can be operated by community health workers, supporting rapid data reporting. After co-designing the implementation package using a human-centered design approach with Uganda’s Ministry of Health, we are implementing a surveillance strategy that incorporates VectorCam into routine work. Objective: This study aims to evaluate acceptability, fidelity, feasibility, cost, and preliminary effectiveness of VectorCam, the intervention, along with supporting implementation strategies as it is integrated into Uganda’s national malaria control program. Methods: This mixed methods hybrid type III implementation-effectiveness study uses district-level restricted randomization to assign 12 districts to four sequential rollout waves following a 2-district pilot phase. The implementation package will be introduced over a one-year period across malaria-endemic districts in Northern and Eastern Uganda. Implementation outcomes, including acceptability, feasibility, fidelity, and cost, will be assessed using qualitative interviews, system usage logs, field observations, and program documentation. Effectiveness outcomes include technical performance of mosquito species identification, time efficiency, and operational use of surveillance data for decision-making. Accuracy will be evaluated through comparison with molecular identification methods, alongside additional classification performance metrics. Data will be analyzed using a mixed methods approach, with triangulation of qualitative and quantitative data and longitudinal analyses across districts and rollout waves. An adaptive learning approach will guide iterative refinement of the implementation package. Results: Fourteen districts were selected by the Ministry of Health in September 2025 for VectorCam introduction. A pilot phase conducted between December 2025 and January 2026 in two districts was used to refine implementation strategies and research tools through stakeholder workshops. Ethical approval was obtained from Johns Hopkins in January 2026 and Ugandan ethics committees in September 2025. Implementation began in February 2026 and will continue through March 2027 with concurrent data collection and analysis throughout the implementation timeline. Conclusions: This study will evaluate a task-shifting approach to entomological surveillance by integrating an AI-enabled tool into routine workflows of community health workers through a co-designed implementation package. By examining how the VectorCam intervention and supporting implementation strategies function together within real-world health system contexts, this study will generate evidence on the sustainability of digitally enabled surveillance. Findings will provide insight into how task shifting and digital tools can expand surveillance capacity, improve timeliness of data reporting, and strengthen the use of entomological data for decision-making. These results will inform national scale-up efforts in Uganda and offer a model for other malaria-endemic settings seeking to modernize vector surveillance.

  • Open Access

    Integration of VectorCam, an AI-Enabled Mosquito Surveillance Tool, into Uganda’s National Malaria Elimination Program: Protocol for a Mixed Methods Hybrid Type III Implementation Study (Preprint)

    by Mei-Li Hey, Marina Rincon-Torroella, Diwakar Mohan, Sunny Patel, Winnifred Kansiime, Aryaman Shodhan, Catherine Maiteki Sebuguzi, Youseph Yazdi, Peter Waiswa, Soumyadipta Acharya

    BACKGROUND Malaria remains one of the most significant global health threats, and its control depends on timely, accurate entomological surveillance. In Uganda, routine vector surveillance relies heavily on microscope-based mosquito identification by a limited number of trained entomologists, which can delay reporting. VectorCam is an artificial intelligence (AI) enabled digital tool to support real-time mosquito identification and can be operated by community health workers, supporting rapid data reporting. After co-designing the implementation package using a human-centered design approach with Uganda’s Ministry of Health, we are implementing a surveillance strategy that incorporates VectorCam into routine work. OBJECTIVE This study aims to evaluate acceptability, fidelity, feasibility, cost, and preliminary effectiveness of VectorCam, the intervention, along with supporting implementation strategies as it is integrated into Uganda’s national malaria control program. METHODS This mixed-methods hybrid type III implementation-effectiveness study uses district-level restricted randomization to assign 12 districts to four sequential rollout waves following a two-district pilot phase. The implementation package will be introduced over a one-year period across malaria-endemic districts in Northern and Eastern Uganda. Implementation outcomes, including acceptability, feasibility, fidelity, and cost, will be assessed using qualitative interviews, system usage logs, field observations, and program documentation. Effectiveness outcomes include technical performance of mosquito species identification, time efficiency, and operational use of surveillance data for decision-making. Accuracy will be evaluated through comparison with molecular identification methods, alongside additional classification performance metrics. Data will be analyzed using a mixed-methods approach, with triangulation of qualitative and quantitative data and longitudinal analyses across districts and rollout waves. An adaptive learning approach will guide iterative refinement of the implementation package. RESULTS Fourteen districts were selected by the Ministry of Health in September 2025 for VectorCam introduction. A pilot phase conducted between December 2025 and January 2026 in two districts was used to refine implementation strategies and research tools through stakeholder workshops. Ethical approval was obtained from Johns Hopkins in January 2026 and Ugandan ethics committees in September 2025. Implementation began in February 2026 and will continue through March 2027 with concurrent data collection and analysis throughout the implementation timeline. CONCLUSIONS This study will evaluate a task-shifting approach to entomological surveillance by integrating an AI–enabled tool into routine workflows of community health workers through a co-designed implementation package. By examining how the VectorCam intervention and supporting implementation strategies function together within real-world health system contexts, this study will generate evidence on the sustainability of digitally enabled surveillance. Findings will provide insight into how task shifting and digital tools can expand surveillance capacity, improve timeliness of data reporting, and strengthen the use of entomological data for decision-making. These results will inform national scale-up efforts in Uganda and offer a model for other malaria-endemic settings seeking to modernize vector surveillance.


  • Open Access

    Towards transforming malaria vector surveillance using VectorBrain: a novel convolutional neural network for mosquito species, sex, and abdomen status identifications

    by Deming Li, Shruti Hegde, Aravind Sunil Kumar, Atul Zacharias, Parthvi Mehta, Venkat Mukthineni, Satwik Srimath, Sunny Patel, Maitreya Suin, Rama Chellappa, et al.

    Malaria is a major public health concern, causing significant morbidity and mortality globally. Monitoring the local population density and diversity of the vectors transmitting malaria is critical to implementing targeted control strategies. However, the current manual identification of mosquitoes is a time-consuming and intensive task, posing challenges in low-resource areas like sub-Saharan Africa; in addition, existing automated identification methods lack scalability, mobile deployability, and field-test validity. To address these bottlenecks, a mosquito image database with fresh wild-caught specimens using basic smartphones is introduced, and we present a novel CNN-based architecture, VectorBrain, designed for identifying the species, sex, and abdomen status of a mosquito concurrently while being efficient and lightweight in computation and size. Overall, our proposed approach achieves 94.44±2% accuracy with a macro-averaged F1 score of 94.10±2% for the species classification, 97.66±1% accuracy with a macro-averaged F1 score of 96.17±1% for the sex classification, and 82.20±3.1% accuracy with a macro-averaged F1 score of 81.17±3% for the abdominal status classification. VectorBrain running on local mobile devices, paired with a low-cost handheld imaging tool, is promising in transforming the mosquito vector surveillance programs by reducing the burden of expertise required and facilitating timely response based on accurate monitoring.


  • Open Access

    A Handheld Tool for the Rapid Morphological Identification of Mosquito Species (VectorCam) for Community-Based Malaria Vector Surveillance: Summative Usability Study

    by Saisamhitha Dasari, Bhavya Gopinath, Carter James Gaulke, Sunny Mahendra Patel, Khalil K Merali, Aravind Sunil Kumar, Soumyadipta Acharya

    Background: Malaria impacts nearly 250 million individuals annually. Specifically, Uganda has one of the highest burdens, with 13 million cases and nearly 20,000 deaths. Controlling the spread of malaria relies on vector surveillance, a system where collected mosquitos are analyzed for vector species’ density in rural areas to plan interventions accordingly. However, this relies on trained entomologists known as vector control officers (VCOs) who identify species via microscopy. The global shortage of entomologists and this time-intensive process cause significant reporting delays. VectorCam is a low-cost artificial intelligence–based tool that identifies a mosquito’s species, sex, and abdomen status with a picture and sends these results electronically from surveillance sites to decision makers, thereby deskilling the process to village health teams (VHTs). Objective: This study evaluates the usability of the VectorCam system among VHTs by assessing its efficiency, effectiveness, and satisfaction. Methods: The VectorCam system has imaging hardware and a phone app designed to identify mosquito species. Two users are needed: (1) an imager to capture images of mosquitos using the app and (2) a loader to load and unload mosquitos from the hardware. Critical success tasks for both roles were identified, which VCOs used to train and certify VHTs. In the first testing phase (phase 1), a VCO and a VHT were paired to assume the role of an imager or a loader. Afterward, they swapped. In phase 2, two VHTs were paired, mimicking real use. The time taken to image each mosquito, critical errors, and System Usability Scale (SUS) scores were recorded for each participant. Results: Overall, 14 male and 6 female VHT members aged 20 to 70 years were recruited, of which 12 (60%) participants had smartphone use experience. The average throughput values for phases 1 and 2 for the imager were 70 (SD 30.3) seconds and 56.1 (SD 22.9) seconds per mosquito, respectively, indicating a decrease in the length of time for imaging a tray of mosquitos. The loader’s average throughput values for phases 1 and 2 were 50.0 and 55.7 seconds per mosquito, respectively, indicating a slight increase in time. In terms of effectiveness, the imager had 8% (6/80) critical errors and the loader had 13% (10/80) critical errors in phase 1. In phase 2, the imager (for VHT pairs) had 14% (11/80) critical errors and the loader (for VHT pairs) had 12% (19/160) critical errors. The average SUS score of the system was 70.25, indicating positive usability. A Kruskal-Wallis analysis demonstrated no significant difference in SUS (H value) scores between genders or users with and without smartphone use experience. Conclusions: VectorCam is a usable system for deskilling the in-field identification of mosquito specimens in rural Uganda. Upcoming design updates will address the concerns of users and observers.


  • Navigating Breast Cancer Innovation in Uganda: Within and Without the Textbook

    by Kim Hwang Yeo, Krishna Tejaswini Sathi, Soumyadipta Acharya, Youseph Yazdi

    The Center for Bioengineering Innovation and Design (CBID) at Johns Hopkins University (JHU) has established a comprehensive approach to addressing global health challenges. Central to CBID’s modality on global health is a strategy that integrates education, research, and collaboration. Through its graduate program, CBID trains the next generation of health care innovators to address the specific needs of low- and middle-income countries (LMICs). Graduate student teams at CBID begin their year with a focus on a health care thematic area associated with a target country.


  • Health Care Innovation for Low-Resource Settings: The Value of Local Immersion and Partnership

    by K. S. Parikh, A. Fuleihan, S. Acharya, T. Sathi, T. Hasan, K. H. Yao, Y. Yazdi

    Health Care Innovation is the creation, development, and translation of new and better solutions to health care challenges. At its core, this endeavor does not require extending the frontiers of science or the creation of new fundamental technologies. Rather, it is primarily focused on the use of existing science and established technologies in the design of new solutions to problems in health care. Successfully innovating for low- and middle-income countries (LMICs) requires a needs and stakeholder-driven approach to enable development and adoption of available, accessible, and acceptable solutions tailored to the specific need and context of care.


  • Open Access

    A Handheld Tool for the Rapid Morphological Identification of Mosquito Species (VectorCam) for Community-Based Malaria Vector Surveillance: Summative Usability Study (Preprint)

    by Saisamhitha Dasari, Bhavya Gopinath, Carter James Gaulke, Sunny Mahendra Patel, Khalil K Merali, Aravind Sunil Kumar, Soumyadipta Acharya

    BACKGROUND Malaria impacts nearly 250 million individuals annually. Specifically, Uganda has one of the highest burdens, with 13 million cases and nearly 20,000 deaths. Controlling the spread of malaria relies on vector surveillance, a system where collected mosquitos are analyzed for vector species’ density in rural areas to plan interventions accordingly. However, this relies on trained entomologists known as vector control officers (VCOs) who identify species via microscopy. The global shortage of entomologists and this time-intensive process cause significant reporting delays. VectorCam is a low-cost artificial intelligence–based tool that identifies a mosquito’s species, sex, and abdomen status with a picture and sends these results electronically from surveillance sites to decision makers, thereby deskilling the process to village health teams (VHTs). OBJECTIVE This study evaluates the usability of the VectorCam system among VHTs by assessing its efficiency, effectiveness, and satisfaction. METHODS The VectorCam system has imaging hardware and a phone app designed to identify mosquito species. Two users are needed: (1) an imager to capture images of mosquitos using the app and (2) a loader to load and unload mosquitos from the hardware. Critical success tasks for both roles were identified, which VCOs used to train and certify VHTs. In the first testing phase (phase 1), a VCO and a VHT were paired to assume the role of an imager or a loader. Afterward, they swapped. In phase 2, two VHTs were paired, mimicking real use. The time taken to image each mosquito, critical errors, and System Usability Scale (SUS) scores were recorded for each participant. RESULTS Overall, 14 male and 6 female VHT members aged 20 to 70 years were recruited, of which 12 (60%) participants had smartphone use experience. The average throughput values for phases 1 and 2 for the imager were 70 (SD 30.3) seconds and 56.1 (SD 22.9) seconds per mosquito, respectively, indicating a decrease in the length of time for imaging a tray of mosquitos. The loader’s average throughput values for phases 1 and 2 were 50.0 and 55.7 seconds per mosquito, respectively, indicating a slight increase in time. In terms of effectiveness, the imager had 8% (6/80) critical errors and the loader had 13% (10/80) critical errors in phase 1. In phase 2, the imager (for VHT pairs) had 14% (11/80) critical errors and the loader (for VHT pairs) had 12% (19/160) critical errors. The average SUS score of the system was 70.25, indicating positive usability. A Kruskal-Wallis analysis demonstrated no significant difference in SUS (H value) scores between genders or users with and without smartphone use experience. CONCLUSIONS VectorCam is a usable system for deskilling the in-field identification of mosquito specimens in rural Uganda. Upcoming design updates will address the concerns of users and observers. CLINICALTRIAL