Artificial intelligence (AI) tools for breast cancer provide decision-supporting models to health care professionals. A research project on breast cancer, in collaboration with Susan G. Komen, was highlighted on the National Science Foundation's (NSF) website.  Click here to read the story on the NSF website.

More than 40,000 women and 500 men in the United States are estimated to die from breast cancer each year, according to the Cancer Facts & Figures published every year by the American Cancer Society. Although the medical field is full of highly skilled and dedicated health care workers, armed with many treatments and care plans for breast cancer patients, the disease remains the second most common cause of death from cancer in women in the United States, according to the U.S. Department of Health and Human Services. However, a novel computer model could help reduce those unacceptable numbers and provide better decision-making models that give practitioners the information they need to discuss tailored treatments with patients.

The artificial intelligence (AI) tool is built using 1.6 million anonymous health care entries from more than 260,000 breast cancer patients over a period of 41 years. Through meaningful analysis of those many data points, including patient demographics and medical actions taken, the model aims to provide physicians with decision-making information that personalizes treatment. The researchers pulled anonymous patient data from public breast cancer databases, such as those run by the National Cancer Institute and Centers for Disease Control to amass 1.6 million entries in all from patients diagnosed and treated between 1975 to 2016.

The entries follow every patient from the time of first diagnosis and through each outcome over the years. Those data come from patients that have already undergone treatment, and they will allow the algorithm to learn from how patients progressed following what actions. With iterative trainings, it will become more accurate at formulating the probability that certain actions will lead to specific disease progressions. Eventually, says Yang, the final model can help Komen with its mission of saving lives, through informing its education programs, public policy advocacy, and support of research to prevent and cure breast cancer.

Yang and his team will also turn the algorithm into a software application that physicians can use to help make decisions and further offer precision oncology to their patients. For instance, a doctor would input details about his patient (demographics and disease characteristics) to receive the probabilistic estimate that each action throughout the patient's tailored treatment plan will lead to the desired disease progression and, ultimately, favorable outcome for that patient.

A Penn State CHOT team, consisting of  Bryant Niederriter  and Alice Rong , along with CHOT faculty adviser Faisal Aqlan,  developed a virtual-reality based method to help the assessment of mental disorders.  Recent reports show that 1 in 4 families has at least one member with a mental disorder. The current diagnosis in psychiatry is based on clinical interviews and questionnaires, which are subjective and can lead to recalls and interviewer biases. Virtual reality (VR) is a new immersive technology that provides  an unprecedented opportunity to measure one's  feedback in the virtual environment. 

VR waiting room of clinical center

In health care, VR shows strong potential to improve decision making and help patients to better connect with reality, cope with pain, and overcome mental disorders such as anxiety and depression. This study integrates sensing technology (i.e., eye tracking) with VR simulation of health care environments to improve clinical judgment and assessment of mental disorders. Traditional scenario-based patient simulations are used as a basis for the development of VR modules. Data collected via sensing technology are utilized to develop analytical models for predicting the risk of mental illness. Moreover, artificial intelligence (AI) tools for VR-based health care training help students learn faster and make smarter decisions. This research helps contribute to improved population health by developing new methods for promoting health and effectively predicting and treating mental disorders. Here is the demo video of the described VR system.

Demo image of the VR system

In the April edition of our CHOT newsletter, we announced that five papers from CHOT scholars were accepted to IEEE EMBC 2020. Due to the COVID-19 pandemic, the organizing committee of the conference decided that for this year, EMBC will be a virtual conference. Below are presentations of the accepted papers by the authors submitted to the conference.

1. Dynamic Physician-Patient Matching in the Health Care System

2. Markov Decision Process Modeling for Multi-stage Optimization of Intervention and Treatment Strategies in Breast Cancer

3. Statistical Analysis of County-Level Contributing Factors to Opioid-Related Overdose Deaths in the United States

4. Gradient Mechanism to Preserve Differential Privacy and Deter Against Model Inversion Attacks in Health Care Analytics

5. Hierarchical Gaussian Process Modeling and Estimation of State-action Transition Dynamics in Breast Cancer