AI: Real Data-Driven Analytical Predictive Model For Pancreatic Cancer, Survival Monitoring Indicator, Survival Methodologies

Tech ID: 22A113

­Competitive Advantages:

  • The inventive model identifies the risk factors, and interaction and predicts the survival time of a pancreatic cancer patient.

  • The analytic survival indicator can be used to evaluate the effectiveness of different treatments given to patients, by identifying if the survival time of the patients is increasing, remaining the same, or decreasing.

  • The innovation includes additional findings concerning significant differences in gender, age, and race of pancreatic cancer patients and identifies appropriate methodologies for more powerful survival predictions of the disease.

  • The findings and predictions about pancreatic cancer have been statistically verified to be 96.4% accurate.

Summary:

One of the deadliest cancers with low survival rates is pancreatic cancer, which is difficult to identify in its early stages and has a very low chance of recovery after it has spread. An AI model that can forecast a patient's survival time has been built by researchers at the University of South Florida. Based on the risk factors involved, which have been identified according to their proportional contribution to the model, the model forecasts this rate. The model's accuracy was determined through statistical testing, and it was found to be 96.42%. A stochastic model and survival monitoring indicator are incorporated into the model to further improve the predictions. The stochastic model tracks the patient's behavior at a given time, whereas the survival monitoring indicator tracks the patient's survival rate as a function of time. The efficacy of the various medical treatments can be evaluated using all these signs and forecasts. The unique feature of this paradigm is that it can include a variety of additional treatments in addition to chemotherapy, radiation, and other common therapies, and clinicians can choose which therapy to use next based on its efficacy. Additionally, the technology allows for the classification of prognosis and therapy based on age and race.

(i) Risk Factors involved  in Pancreatic Cancer

(ii) Histogram, CDF, and Probability Density of Survival Times of Male Pancreatic Cancer Patients in different Stages.

Desired Partnerships:

  • License

  • Sponsored Research

  • Co-Development

Technology Transfer
TTOinfo@usf.edu
(813) 974-0994

Patents