At Purkayastha Laboratory, one of our research focal points is in the area of improving Radiological Information Systems. We believe healthcare should be patient-centric and one of the ways is to ensure that patients are actively involved in their care. Our vision includes improving patient awareness especially in the area of radiology. That is why we have all hands on deck working relentlessly towards extensive research on improvement Patient-Provider-Radiologist Communication.
We have been awarded a grant of $826,795.00 on Measuring learning gains in man-machine assemblage when augmenting radiology work with artificial intelligence by the National Science Foundation(NSF).
The project will answer the following research questions: Q1: How to develop assemblages, such that human-technology partnerships produce a “good fit” for visually based cognition-oriented tasks in radiology? Q2: What level of training should pre-exist in the individual human (radiologist) and independent machine learning model for human-technology partnerships to thrive? Q3: Which aspects and to what extent does an assemblage learning approaches lead to reduced errors, improved accuracy, faster turn-around times, reduced fatigue, improved self-efficacy, and resilience?
Over the years, there have been various research on patient education and radiology. This research work was conducted by notable research who saw the need for two-way communication between a Physician and patients. Our aim is to feed the cognitive ability of patients to easily interpret images without any form of formal education in medicine.
Our various research projects are based on past research works with strong footing and we also endeavor to engage in surveys and all other forms of inquiry to ensure patients are actively involved in the design of Radiological Systems. in those ways, we believe we can work towards more accurate diagnoses, fewer medical errors, and improved complete communication among all parties involved.
- Cognitive Fit:
By using the Cognitive fit Theory, which has been used to study interpretation differences between people with varied cognitive abilities, there is Imaging 3.0, which is an initiative led by the American College of Radiology (ACR) to make radiologists more participative inpatient care. This is by establishing collaborative working processes between Radiologist, Patient, and Referring physician (RP). One of the envisaged goals of the initiative is “to deliver all the imaging care that is beneficial and necessary and none that is not.” Among the different parts of Imaging 3.0, awareness through advocacy, economics, education, quality and safety, and clinical research are important pillars of the ACR (Haines, 2013). With these pillars in mind, we see the need to research the challenges and opportunities that are presented in collaborative work between the actors such as radiologists, patients, RP, and technology, including its underlying information – images, videos etc. We argue that due to the differences in the cognitive abilities of radiologists, patients, and RP, the way in which radiology images are displayed to each Use group should be articulated differently.
Our primary objective is to evaluate the difference in the cognitive abilities of these user groups when looking at and interpreting radiology images such as X-rays, MRI, CT, and USG. The study does not focus on the generalizability of the findings to specific statistical measures. Rather it is an attempt to describe the individual abilities (and then extrapolate to the groups of users) in the context of their work in interpreting radiology studies with their own set of cognitive abilities. The factors identified in cognitive fit theory will be correlated using Partial Least Squares (PLS) regression between the factors. The PLS model will highlight the different abilities of the users and how they correlate between internal problem representation, understanding of the application domains, tasks performed for problem-solving, and the mental presentation of the task solution.
References:
Using Cognitive fit to analyze challenges and opportunities in Imaging 3.0, Dr. Saptarshi Purkayastha.
- Human Vs Machine Project:
We have developed a machine-learning algorithm to diagnose Pneumonia by analyzing the X-rays. We will run that algorithm on some Radiological systems and collect the Machine-generated results. In the same way, we shall collect inferences of Radiological systems from the Radiologist to generate Human/User-generated results. By comparing the user-generated results and machine-generated results, the user will get a score against the machine. Using this data, an inference is drawn using Python, HTML, JavaScript, and SQL database.
- Radiological Information systems – Beta:
Radiological Information Systems can greatly improve patient care and physician education. However, these features are not being accessed by most of the developing countries due to its high-cost constraints. By using these Beta Radiological systems, our research team is in the process of exploring low-cost options for basic RIS. In this process, we installed a teaching hospital in a developing country and used it until hardware failure six months after its installation. During this time, nearly 2000 studies had been stored in a database. We utilized the Institution’s existing internet connection to provide e-mail correspondence with radiologists at the Indiana University.
One of the major difficulties facing any system is user acceptance. Our system has been a hit among the users since the beginning. We continue to receive emails from many countries requesting the continuation of the project. This, we consider as an unequivocal proof-of-concept.
We are now developing an open-source solution built with a robust data model, existing libraries, high-level development tools, and standards-based network communication. We are integrating the image viewer(OHIF) and a PACS system (Orthanc) with LibreHealth Radiology in this release. This second-generation system will support customized web-based or stand-alone clients.
- Consumer Perspective and Patient Education of Radiological Images
We are currently running a survey that is targeted toward patients who have recently undergone any form of a radiological scan. We believe their perception would better inform us to develop a solution suitable for them. By ensuring coordination between our technical term members and our clinical focused researchers we believe we would come up with the next generation of integrated Radiological Information Systems.
- Radiological Module for OpenMRS systems
Purkayastha laboratory has also been actively involved in the development of OpenMRS, an open-source Medical Records System designed to cater for the need of the developing countries. OpenMRS has been deployed in many countries around the world and it’s still gaining recognition on a daily basis.