Team: 8
School: Clovis High Early College
Area of Science: Medicine
Proposal: Cancer is the second-leading cause of death worldwide. There are many possible methods to treat Cancer. Traditionally, to make a patient’s cancer treatment plan, information often needs to be utilized among many different specialist practitioners, including a pathologist, surgeon, medical oncologist, and more depending on the case. It can be an elaborate process. We plan to introduce a method capable of compiling Cancer treatment plans with a higher rate of potency using machine learning. We will achieve this with a program that has a Python backend and a Java frontend. Our training and testing data will mainly be sourced from the Cancer Imaging Archive, a publicly available resource with a wide range of Cancer data. Since the amount of data a patient corresponds with changes from case to case, the model will be scalable and work with a number of data dimensions. We plan to make two versions of our program: one that utilizes classical machine learning technology and utilizes the up and coming technology of Quantum Machine Learning (QML). QML has become very accessible via IBM’s Quantum Experience program and Python package Qiskit. Using these resources, we can access real Quantum Computers to run our programs on for free. Since, at the moment, it is not possible to make a fully quantum neural network, our QML program will have a hybrid design with the input and output layers having a classical system, and the middle layer utilizing a quantum architecture. The intermediate layers will run on IBM’s quantum computers over the cloud while the input and output layers will run on the local system. In addition to our primary goal, we also hope to demonstrate some form of quantum superiority over the classical system.
Mentor: Alan Daugherty
Team Members:
Erynn Vetterly
Alexis Brandsma
Tristen Pool
Sponsoring Teacher: Alan Daugherty