Med student honored for sickle cell disease project

Aggarwal says he was “definitely surprised” when he learned of his award. “When I started the project, I was not sure if I would even have an end product that would be useful. I think the award really shows how in the next few years there will be significant integration of artificial intelligence with medical management and treatment, especially since health care is one of the few, if not the only, industry that has not had a significant technological revolution.”

More than 100,000 Americans have sickle cell disease, which causes shortened life expectancy, multiple complications and acute pain. The objective of the project was to identify risk factors that significantly impact mortality in hospitalized sickle cell disease (SCD) patients and to develop a model to predict the mortality risk in future patients. The analyzed risk factors included demographic variables, hospital-related data and co-morbidities, such as cardiovascular diseases, infectious diseases, mental disorders and musculoskeletal diseases.

“Nine artificial intelligence machine learning and deep-learning classification systems were used to develop predictive models using the risk factors to quantify the mortality risk in hospitalized SCD patients,” Aggarwal explains. “The output of the model with the highest accuracy can be used by the medical team to allocate more resources and be more aggressive in management of SCD patients with a higher risk of mortality. Use of these models has the potential to reduce mortality in SCD patients as the medical team is provided with objective data that can guide medical management of the patient.”

Aggarwal adds that as new data is obtained daily from hospital admissions, these artificial intelligence models can be easily updated with the new data to further improve accuracy and predictability.

Aggarwal plans to work with hospitals and medical organizations to add more variables and input more data for the current variables so that the model’s accuracy can further improve.

“I would like to also run the machine learning models in a pilot program at a medical care facility in order to assess its real-world impact. Hopefully, this model can serve as a guide for developing models to assess mortality risk in other diseases as well.”

The Healthathon award also will provide Aggarwal with design thinking coaching and mentorship from representatives of AARP and Healthbox, a health care-focused accelerator which assisted with judging.

Aggarwal wants to translate data analytics and the artificial intelligence techniques to applications in dermatology. He has been applying similar principles of artificial intelligence in dermatology with image recognition of skin lesions. He also hopes for an opportunity to impact treatment modalities for hemoglobinopathies because he has Beta Thalassemia trait and would like to do research on diseases for which effective and affordable treatment options have yet to be developed.

“I am fascinated by the applicability of artificial intelligence in health care,” he says. “There is such a vast amount of tabular and graphical data that is generated in health care; yet, much of this data has not been thoroughly analyzed and is essentially wasted. I came across this healthathon and was intrigued by how even though sickle cell disease was first described in 1910, we have still not been able to find an effective, safe and affordable cure for it. I wanted to see if the latest techniques in machine learning and deep learning could be applied to improve the medical management of SCD patients who present to the hospital.”