Saptarshi Purkayastha
Permanent URI for this collection
Risks and Opportunities of AI Recognition of Patient Race in Medical Imaging
Previous studies in medical imaging have shown disparate abilities of artificial intelligence (AI) to detect a person's race, yet there is no known correlation for race on medical imaging that would be obvious to human experts when interpreting the images. His recent work published in Lancet Digital Health demonstrates that deep learning models have extremely high accuracy at identifying self-reported race from medical images such as X-rays, MRIs and CTs. This ability raises serious concerns among some researchers. Such software might group patients, or influence their care, by factoring in race. These AI models work very well on poor quality, distorted and even images where many parts of the image were deliberately cut out. These types of categorizations could lead to inequality in providing health care and making recommendations, and human decision makers might not understand how and why AI models are making the recommendations. Engineers, clinical researchers and informaticians need to get together to identify how AI models are able to have these superhuman capabilities.
Professor Purkayastha's translation of research into potential ways to identify and mitigate risks of deploying AI models in clinical practice to avoid racial issues in healthcare treatment is another example of how IUPUI's faculty members are TRANSLATING their RESEARCH INTO PRACTICE.
Browse
Browsing Saptarshi Purkayastha by browse.metadata.dateaccessioned
Now showing 1 - 10 of 64
Results Per Page
Sort Options
Item Overview, not Overwhelm: Framing Operational BI Tools using Organizational Capabilities(2013) Purkayastha, Saptarshi; Braa, JørnIn contexts where fragmentation of information systems is a problem, data warehouse (DW) has brought disparate sources of information together. While bringing data together from multiple health programs and patient record systems, how does one make sense of huge amounts of integrated information? Recent research and industry uses the term, “Operational BI” for decision making tools used in operational activities. In this paper, we highlight the use of DHIS 2, a large-scale, open-source, Health Management Information System (HMIS) that acts as a DW. Firstly, we present the results of a survey done in 13 countries to assess how Operational BI Tools are used. We then show 3 generations of BI Tools in DHIS 2 that have evolved from action-research done over 18 years in more than 30 countries. Secondly, we develop the Overview-Overwhelm (O-O) analytical framework for large-scale systems that need to work with Big Data. The O-O framework combines lessons from DHIS 2 BI Tools design and implementation survey results.Item OpenScrum: Scrum methodology to improve shared understanding in an open-source community(2014) Purkayastha, SaptarshiWhile we continue to see rise in the adoption of agile methods for software development, there has been a call to study the appropriateness of agile methods in open-source and other emerging contexts. This paper examines Scrum methodology adopted by a large, globally distributed team which builds an open-source electronic medical records platform called OpenMRS. The research uses a mixed method approach, by doing quantitative analysis of source-code, issue tracker as well as community activity (IRC logs, Mailing lists, wiki) in pre and post Scrum adoption, covering a period of 4 years. Later we conducted semi-structured interviews with core developers and followed it up with group discussions to discuss the analysis of the quantitative data and get their views on our findings. Since the project is "domain heavy", contributors (developers and implementers) need to have certain health informatics understanding before making significant contributions. This puts knowledge-sharing and "bus factor" as critical points of management for the community. The paper presents ideas about a tailored Scrum methodology that might better suited for open-source communities to improve knowledge-sharing and community participation, instead of just agilityItem Designing a drawing-based tool to manage EBRT process in an open-source oncology EMR system(AMIA Symposium 2015, 2015-11-14) Maheshwari, Manika; Purkayastha, SaptarshiThis paper describes the community-based participatory research to implement open source Oncology EMR for radiation practices. This tool facilitates better communication between Oncologist, Technician and Patient. The innovation is the use of a Drawing module embedded within the EMR system through which the Radiology technician can visualize medical images.Item Mobile-Application Based Cognitive Behavior Therapy (CBT) for Identifying and Managing Depression and Anxiety(Springer, 2017-07) Addepally, Siva Abhishek; Purkayastha, Saptarshi; BioHealth Informatics, School of Informatics and ComputingMobile technology is a cost effective and scalable platform for developing a therapeutic intervention. This paper discusses the development of a mobile application for people suffering with depression and anxiety. The application which we have developed is similar to a Cognitive Behavior Therapy (CBT) website, which is freely available on the internet. Past research has shown that CBT delivered over the internet is effective in alleviating the depressive symptoms in users. But, this delivery method is associated with some innate drawbacks, which caused user dropout and reduced adherence to the therapy. To overcome these shortfalls, from web based CBT delivery, a mobile application called MoodTrainer was developed. The application is equipped with mobile specific interventions and CBT modules which aim at delivering a dynamic supportive psychotherapy to the user. The mobile specific interventions using this application ensures that the user is constantly engaged with the application and focused to change the negative thought process. We present MoodTrainer as a self-efficacy tool and virtual CBT that is not meant to replace a clinical caregiver. Rather, it is a supportive tool that can be used to self-monitor, as well as a monitoring aid for clinicians.Item Conversion of JPG Image into DICOM Image Format with One Click Tagging(Springer, 2017) Oladiran, Olakunle; Gichoya, Judy; Purkayastha, Saptarshi; BioHealth Informatics, School of Informatics and ComputingDICOM images are the centerpiece of radiological imaging. They contain a lot of metadata information about the patient, procedure, sequence of images, device and location. To modify, annotate or simply anonymize images for distribution, we often need to convert DICOM images to another format like jpeg since there are a number of image manipulation tools available for jpeg images compared to DICOM. As part of a research at our institution to customize radiology images to assess cognitive ability of multiple user groups, we created an open-source tool called Jpg2DicomTags, which is able to extract DICOM metadata tags, convert images to lossless jpg that can be manipulated and subsequently reconvert jpg images to DICOM by adding back the metadata tags. This tool provides a simple, easy to use user-interface for a tedious manual task that providers, researchers and patients might often need to do.Item Comparative Performance Analysis of Different Fingerprint Biometric Scanners for Patient Matching(IOS Press, 2017) Kasiiti, Noah; Wawira, Judy; Purkayastha, Saptarshi; Were, Martin C.; BioHealth Informatics, School of Informatics and ComputingUnique patient identification within health services is an operational challenge in healthcare settings. Use of key identifiers, such as patient names, hospital identification numbers, national ID, and birth date are often inadequate for ensuring unique patient identification. In addition approximate string comparator algorithms, such as distance-based algorithms, have proven suboptimal for improving patient matching, especially in low-resource settings. Biometric approaches may improve unique patient identification. However, before implementing the technology in a given setting, such as health care, the right scanners should be rigorously tested to identify an optimal package for the implementation. This study aimed to investigate the effects of factors such as resolution, template size, and scan capture area on the matching performance of different fingerprint scanners for use within health care settings. Performance analysis of eight different scanners was tested using the demo application distributed as part of the Neurotech Verifinger SDK 6.0.Item Improving “Desktop medicine” efficiency using Guided Inquiry Learning in an Electronic Health Records System(2018-07-18) Purkayastha, Saptarshi; Naliyatthaliyazchayil, Parvati Ravindranathan Menon; Surapaneni, Asha Kiranmayee; Kowkutla, Ashwini; Maity, PallaviRecent studies have shown that more than 50% of physician time is spent on “desktop medicine” – the practice of medicine that requires the use of computers and other technology. Providers are trained in other medical practices through elaborate course work, but don’t get enough training and follow-up training on the desktop medicine practices such as efficient use of an electronic health record (EHR) system. By putting in practice theories from learning sciences, human-computer interaction and evaluation in an undergraduate health information management (HIM) course, we developed a module called Student Team Learning Monitor (STLM) in an open-source EHR.Item A Platform for Innovation and Standards Evaluation: a Case Study from the OpenMRS Open-Source Radiology Information System(Springer, 2018-05-10) Gichoya, Judy W.; Kohli, Marc; Ivange, Larry; Schmidt, Teri S.; Purkayastha, Saptarshi; Radiology and Imaging Sciences, School of MedicineOpen-source development can provide a platform for innovation by seeking feedback from community members as well as providing tools and infrastructure to test new standards. Vendors of proprietary systems may delay adoption of new standards until there are sufficient incentives such as legal mandates or financial incentives to encourage/mandate adoption. Moreover, open-source systems in healthcare have been widely adopted in low- and middle-income countries and can be used to bridge gaps that exist in global health radiology. Since 2011, the authors, along with a community of open-source contributors, have worked on developing an open-source radiology information system (RIS) across two communities-OpenMRS and LibreHealth. The main purpose of the RIS is to implement core radiology workflows, on which others can build and test new radiology standards. This work has resulted in three major releases of the system, with current architectural changes driven by changing technology, development of new standards in health and imaging informatics, and changing user needs. At their core, both these communities are focused on building general-purpose EHR systems, but based on user contributions from the fringes, we have been able to create an innovative system that has been used by hospitals and clinics in four different countries. We provide an overview of the history of the LibreHealth RIS, the architecture of the system, overview of standards integration, describe challenges of developing an open-source product, and future directions. Our goal is to attract more participation and involvement to further develop the LibreHealth RIS into an Enterprise Imaging System that can be used in other clinical imaging including pathology and dermatology.Item Implementing clinical practice guidelines for chronic obstructive pulmonary disease in an EHR system(IEEE, 2017-11) Walker, Marisa; Ge, WeiWei; Gichoya, Judy W.; Purkayastha, Saptarshi; BioHealth Informatics, School of Informatics and ComputingThe use of clinical practice guidelines to improve quality of care has been a vividly discussed topic. Clinical practice guidelines (CPG) aim to improve the health of patients by guiding individual care in clinical settings. CPGs bring potential benefits for patients by improving clinical decision making, improving efficiency and enhancing patient care, while essentially optimizing financial value. Chronic conditions like heart disease, stroke, and chronic obstructive pulmonary disease (COPD), plague the US healthcare system causing several million dollars in healthcare related cost. This paper demonstrates the development of a CPG into an open-source EHR system to effectively manage COPD patients. The CPG is incorporated using the open web app standard, which allows it to be used with any web browser based EHR system, once data from the EHR system can be fed into the app. As a result, the CPG helps create a more effective and efficient decision-making process.Item Evaluating the Implementation of Deep Learning in LibreHealth Radiology on Chest X-Rays(Springer, 2019-04-25) Purkayastha, Saptarshi; Buddi, Surendra Babu; Yadav, Bhawana; Nuthakki, Siddhartha; Gichoya, Judy W.Respiratory diseases are the dominant cause of deaths worldwide. In the US, the number of deaths due to chronic lung infections (mostly pneumonia and tuberculosis), lung cancer and chronic obstructive pulmonary disease has increased. Timely and accurate diagnosis of the disease is highly imperative to diminish the deaths. Chest X-ray is a vital diagnostic tool used for diagnosing lung diseases. Delay in X-Ray diagnosis is run-of-the-mill milieu and the reasons for the impediment are mostly because the X-ray reports are arduous to interpret, due to the complex visual contents of radiographs containing superimposed anatomical structures. A shortage of trained radiologists is another cause of increased workload and thus delay. We integrated CheXNet, a neural network algorithm into the LibreHealth Radiology Information System, which allows physicians to upload Chest X-rays and identify diagnosis probabilities. The uploaded images are evaluated from labels for 14 thoracic diseases. The turnaround time for each evaluation is about 30 seconds, which does not affect clinical workflow. A Python Flask application hosted web service is used to upload radiographs into a GPU server containing the algorithm. Thus, the use of this system is not limited to clients having their GPU server, but instead, we provide a web service. To evaluate the model, we randomly split the dataset into training (70%), validation (10%) and test (20%) sets. With over 86% accuracy and turnaround time under 30 seconds, the application demonstrates the feasibility of a web service for machine learning based diagnosis of 14-lung pathologies from Chest X-rays.