1191P - Efficient lung cancer stage prediction and outcome informatics with Bayesian deep learning and MCMC method

Background

Currently, the investigative set for lung cancer primarily comprises of radiological imaging modalities, chest radiographs (X-rays) and CT scans. An increasing number of evidence underscores the pivotal role of screening programs, particularly among high-risk populations, in the early detection of lung cancer. This assertion is substantiated by the findings of three landmark clinical trials, the NLST, the NELSON trial and the UK UKLS. Notably, the employment of low-dose computed tomography (LDCT) has demonstrated remarkable efficacy, exhibiting heightened sensitivity in the detection of early-stage lung neoplasms. Nevertheless, there has not been a development of a robust, structured screening protocol.In this arena, artificial intelligence (AI) emerges as a promising and potent adjunctive tool. AI systems have shown the capacity to augment the precision of early detection of lung malignancies.

Methods

Our research culminated in the development of a Bayesian Neural Network model tailored for lung cancer detection, achieving an accuracy of 99%, thus signaling its potential as a leading-edge diagnostic tool. Our meticulous incorporation of the Hamiltonian Monte Carlo technique ensures precision in exploring the model’s parameter space, with strong credibility and efficacy.

Results

Moving beyond traditional methodologies, our findings not only set new benchmarks in AI-empowered medical diagnostics but also highlight the path for future work aiming at early and accurate cancer detection. Our work distinctly demonstrates, for the first time that the Bayesian Neural Networks (BNNs) can offer a compelling advantage to medical aspects. By providing not only precise predictions but also a clear measure of the associated uncertainty, the BNNs can play a transformative role in lung cancer diagnosis. The superior performance of the BNN model in our experiments, particularly its 99% accuracy accentuates its potential as a diagnostic tool.

Conclusions

In this study, the DNN showcased its robustness with an impressive 93% accuracy rate in predicting lung cancer levels. The architecture’s multi-layered complexity enables it to capture intricate patterns in the data, contributing to its high performance.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.