Establishing a Highly Accurate Circulating Tumor Cell Image Recognition System for Human Lung Cancer by Pre-Training on Lung Cancer Cell Lines

TL;DR

This study creates an AI system that accurately detects lung cancer cells in blood using pre-training on cell line images, reducing the need for large clinical datasets.

Contribution

A novel AI system for CTC detection using pre-training on lung cancer cell lines and transfer learning with limited clinical data.

Findings

Pre-training on lung cancer cell lines significantly improved CTC classification accuracy.

The model achieved 99.51% accuracy with only 17 clinical CTC images.

The method reduces manual effort and improves automation in CTC detection.

Abstract

Circulating tumor cells (CTCs) are rare cancer cells in the blood that can help predict treatment outcomes. However, identifying them manually is slow and needs expertise. In this study, we developed an AI system that accurately detects CTCs using image recognition. To solve the problem of limited clinical images, we first trained the AI system with lung cancer cell line images and then applied transfer learning using a small number of real CTC images. This approach significantly improved accuracy, even with only 17 clinical images. The final model reached 99.5% accuracy. This method reduces the need for large clinical datasets and supports faster, more reliable CTC detection in lung cancer. It may also be applicable to other cancer types and diagnostic workflows.

Background/Objectives: Circulating tumor cells (CTCs) are important biomarkers for predicting prognosis and evaluating treatment efficacy in cancer. We developed the “CTC-Chip” system based on microfluidics, enabling highly sensitive CTC detection and prognostic assessment in lung cancer and malignant pleural mesothelioma. However, the final identification and enumeration of CTCs require manual intervention, which is time-consuming, prone to human error, and necessitates the involvement of experienced medical professionals. Medical image recognition using machine learning can reduce workload and improve automation. However, CTCs are rare in clinical samples, limiting the training data available to construct a robust CTC image recognition system. In this study, we established a highly accurate artificial intelligence-based CTC recognition system by pre-training convolutional neural networks using images from lung cancer cell lines. Methods: We performed transfer learning of convolutional neural networks. Initially, the models were pre-trained using images obtained from lung cancer cell lines. The model’s accuracy was improved by training with a limited number of clinical CTC images. Results: Transfer learning significantly improved the CTC classification accuracy to an average of 99.51%, compared to 96.96% for a model trained solely on pre-trained cell lines (p < 0.05). This approach showed notable efficacy when clinical training images were limited, achieving statistically significant accuracy improvements with as few as 17 clinical CTC images (p < 0.05). Conclusions: Overall, our findings demonstrate that pre-training with cancer cell lines enables rapid and highly accurate automated CTC recognition even with limited clinical data, significantly enhancing clinical applicability and potential utility across diverse cancer diagnostic workflows.

Figures