Developing a Model for Predicting Lung Cancer Using Variational Quantum-Classical Algorithm: A Survey

Document Type



Internal Medicine (East Africa)


There are real life problems that currently proved hard for classical computers (even the best of supercomputers) to solve-the so called computationally intractable problems. Computers and computing devices are limited, by virtue of the fact that they cannot perform certain complex problems. For example, modern cryptography assumes that it is impossible to factorize a large number as the complexity of solving this problem increases exponentially. Even the best supercomputers of today cannot sufficiently find the prime factors of a number with 700-1,000 digits. With quantum computing, however, Shor’s algorithm proved that in principle, a quantum computer can be used to break the security of conventional cryptosystems. Quantum computing also promises exponential improvements for many optimization problems. Lov Grover algorithm was also designed to search for an element in unstructured database. Effort is currently on to leverage quantum computing and quantum mechanical phenomena to tackle complex machine learning problems. Using machine learning algorithm to solve problems is common place. What is however new is using quantum machine learning algorithm to solve problems such as detecting, classifying and predicting disease such as lung cancer. In this work, we highlight the limitations of classical computers to solve certain problems and then propose a hybrid model called the variational quantum-classical algorithm to predict the possibility of a patient developing lung cancer given a set of features which are spelt out in the dataset. The dataset is initially pre-processed, and made to pass through a Parameterized Quantum Circuit (PQC) and then post-processed. Finally, the output of the post-processing phase is used for prediction of lung cancer.


Journal of Applied Artificial Intelligence