Photo: Sean Anthony Eddy/Getty Images
In Himss25, experts will share the details of the algorithm of a well -known PCCI (KTP) patient, which uses the machine learning method without supervision called clustering to identify groups of people based on similar patterns of the apply and access of healthcare, and not just the disease status, according to Jusuf Torera, PhD, main data and scientific scientific data for PCCI.
PCCI is the Parkland Center for Clinical Innovation in Dallas, Texas, Non -Profit of Health Care.
“Know your patient: AI/ml Clustering of diabetes/hypertension’s population” takes place on Thursday, March 6, from 2-3:00 in Venetian | Level 2 | Veronese 2501, at Himss25 in Las Vegas.
In the Dallas network population, the algorithm identifies patients’ clusters with diabetes and hypertension with a combination of social and clinical risk factors.
According to Tamer, gloss analysis discovers basic risk factors, such as the commitment of the judiciary in criminal matters and fears of immigration.
“Identification of basic risk factors, such as the involvement of the judiciary in criminal matters and immigration fears, ensures a comprehensive understanding of the socio-demographic features of patients and potential factors of non-tax factors of the patterns of the use of health services,” Tamier said. “These basic factors are difficult to discover the use of routine questionnaires for obvious reasons, but they appear as basic factors in the analysis of clusters.”
He said that these social factors also contribute to translation problems and other fluctuations culturally to access to healthcare. Using this information, Parkland can enhance the health strategy of the population, ensuring that interventions are more true to meet the needs of patients with diabetes and hypertension.
For example, the scope of the community under certain conditions perceived as “safe” or “friendly” can be more effective in engaging some patients than a digital approach to health or office.
Additional in -depth analyzes identify the missing and potential possibilities of involvement of care that inform about the modifications of work flow using both conventional, based on EHR and non -traditional methods, such as teens and mobile units.
This approach, used to patients with diabetes and hypertension in Dallas, emphasizes the importance of considering factors outside the disease category, such as demographic data, apply, payers’ class, digital involvement and other social health determinants, said Tamer.
“Cooperation of patients in a high -degree of similarity in the field of clinical use, personal and behavioral features allows us to understand the unique medical and social challenges, which are facing each group and the way they affect access to high -quality care and health results,” said Tamer. “This understanding allows the development of targeted clinical programs that are adapted to the specific needs of these patients, ultimately improving patients’ results and allocation of resources.”
Data sets and analytical approaches are scalable and repeatable to other sensitive populations throughout the country.