Exploring the reuse of clinical data for advanced AI pharma research purposes
Accessing medical data for research purposes can be exceedingly challenging, just to mention privacy concerns and regulatory restrictions. Striking a balance between protecting patient confidentiality and enabling valuable research is a complex and ongoing dilemma in the healthcare field. However, as a company providing custom algorithms development services with a research team onboard, we believe in an obligation to maximize the value and utility of this resource.
Furthermore, the reuse of clinical data sets can provide significant benefits to both patients and sponsors. The value of reusing clinical data has been recognized for decades [1]. This approach has demonstrably improved healthcare outcomes and reduced costs. As a result, the pharmaceutical industry continues to show increasing interest in leveraging this valuable resource.
Clinical data reuse – accelerating clinical research
Without a doubt, using clinical trial data in drug discovery boosts the process of discovering and advancing new improved treatments. In consequence it improving patients’ lives ans benefit society as a whole.
Indeed clinical data reuse holds immense potential for accelerating clinical research. Firstly, it allows for expedited patient recruitment, a major bottleneck in many trials. Secondly, this approach facilitates hypothesis testing, enabling researchers to explore new ideas more efficiently. Finally, data reuse provides cost-effective access to a broader range of clinical information, allowing for more comprehensive research applications and feasibility studies on using historical data in new trials. But what are the other goals you may achieve by clinical data reuse or secondary use?
Goals you may achieve by secondary use
The European Federation of Pharmaceutical Industries and Associations (EFPIA) in its framework for secondary use of clinical data [2] provides examples of health research purposes for which clinical trial data might be re-used:
to explore novel hypotheses that would otherwise necessitate launching a new research study involving medical interventions on patients. These fresh hypotheses might focus on investigating alternative treatments or enhancing our comprehension of disease mechanisms, among other possibilities
for regulatory purposes, such as conducting safety studies at the request of authorities and for research (often done in collaboration) involving the pooling of clinical trial data from multiple studies to support in treatment evaluation and drug development.
to enhance the efficiency, design, and methods of future clinical trials as well as to allow independent researchers to validate or scrutinize the original results
and last but not least: to create and test new healthcare technologies, such as AI-based algorithms.
In particular, in pursuit of this final objective, we bring extensive expertise to the table. Throughout the years, we’ve refined our skills and deepened our knowledge in utilizing clinical data to propel the creation of state-of-the-art AI solutions.
Clinical data reuse to create and test AI-based algorithms
We possess a wealth of extensive expertise and knowledge, particularly when it comes to the aforementioned objective. With a proven track record of success in the field, we’re excited to collaborate with like-minded partners and continue our mission of pioneering AI-driven innovations that have a tangible impact and tackle complex challenges. As evidenced by projects described in scientific papers, we have a proven track record in this area.
Our publications
Detecting liver cirrhosis in computed tomography scans using clinically-inspired and radiomic features, Computers in Biology and Medicine.
Krzysztof Kotowski, Damian Kucharski, Bartosz Machura, Szymon Adamski, Benjamín Gutierrez Becker, Agata Krason, Lukasz Zarudzki, Jean Tessier, Jakub Nalepa.
Our team introduced a complete algorithm for the automated detection of cirrhosis using CT. Furthermore, the algorithm benefits from incorporating both clinically inspired and radiomic features.
Deep learning automates bidimensional and volumetric tumor burden measurement from MRI in pre- and post-operative glioblastoma patients.
Jakub Nalepa, Krzysztof Kotowski, Bartosz Machura, Szymon Adamski, Oskar Bozek, Bartosz Eksner, Bartosz Kokoszka, Tomasz Pekala, Mateusz Radom, Marek Strzelczak, Lukasz Zarudzki, Agata Krason, Filippo Arcadu, Jean Tessier.
We developed an algorithm to streamline tumor evaluation using magnetic resonance imaging (MRI). In the first step, this algorithm identifies distinct tumor sub-regions, including the enhancing tumor, peritumoral edema, and surgical cavity. Following the identification of sub-regions, the algorithm calculates volumetric and bidimensional measurements based on the current Response Assessment in Neuro-Oncology (RANO) criteria.
AI algorithms and clinical data-driven innovations
Reusing clinical data to create and test new healthcare technologies, especially AI-based algorithms, has the potential to improve patient care, increase healthcare efficiency, and advance medical research. Are you ready for secondary use of the data to drive groundbreaking research and innovations? Our research team is comprised of experts from diverse fields – data scientists, clinicians, engineers, and more. We are not just developing algorithms; we have experience in applying them in real clinical settings.
If you’re excited about the prospect of reusing clinical data and powering it with AI, reach out. Your team’s expertise, combined with our technological know-how, will be a dynamic force for change. Let’s design and fine-tune AI-based algorithms together to predict, diagnose, or recommend treatments for various medical conditions.
Learn more about our work: Radiology AI Landscape in numbers (and money)
References:
[1] Safran C, Bloomrosen M, Hammond WE, Labkoff S, Markel-Fox S, Tang PC, Detmer DE, Expert Panel. Toward a national framework for the secondary use of health data: an American Medical Informatics Association White Paper. J Am Med Inform Assoc. 2007 Jan-Feb;14(1):1-9 https://doi.org/10.1197%2Fjamia.M2273. Epub 2006 Oct 31. PMID: 17077452; PMCID: PMC2329823.