Towards a More Personalized Approach to Weight Loss
New weight loss drugs, including semaglutide (Ozempic; Wegovy) and tirzepatide (Mounjaro; Zepbound), have revolutionized the treatment of obesity. However, while these drugs help many people lose a significant amount of weight, the results vary widely from person to person. Researchers have developed a genetic test that shows promise in identifying who is most likely to benefit from the newer, more expensive weight loss drugs, and who may respond better to older, less costly—but still effective—drugs. In this pilot project, scientists are working to validate the genetic test by using data from the KP Research Bank. They will look at individuals who have used weight loss drugs to see if the test predicts weight loss. The goal is to make obesity treatment more personalized. If we can determine who responds best to which medications, patients and their care providers can make more informed choices—leading to better outcomes and lower healthcare costs.
Early detection of pancreatic cancer
Pancreatic cancer is the third-leading cause of cancer death. By 2030, scientists predict it will become the second-leading cause of cancer death. One of the main reasons this disease is so deadly is because it’s often not diagnosed until a very late stage. Survival would be greatly improved if people could be diagnosed when the disease is in its early stages.
The scientists leading this project have already analyzed the information from Kaiser Permanente electronic health records related to pancreatic cancer. They used a type of computer programming called “machine learning.” Machine learning is a process by which computers can be given the capability to “learn” about a given dataset — in this case, people who have had pancreatic cancer.
In this new study, the scientists will build on what they’ve already learned from their machine-learning project by developing the next level of machine learning called “deep learning.” Deep learning means that computers can recognize patterns in large datasets and make predictions about the real world without requiring additional input from researchers. This new deep-learning model will use information from KP Research Bank participants’ electronic health records, including results of blood tests, to predict the risk of a person developing pancreatic cancer. This project will also highlight the KP Research Bank as a valuable resource for advancing research on early detection of pancreatic cancer.
