use of AI in cancer screening

Introduction

The use of Artificial Intelligence (AI) in cancer screening has garnered significant interest due to its potential to enhance early detection, improve diagnostic accuracy, and reduce the workload of healthcare professionals. This synthesis aims to consolidate findings from multiple research papers on the application of AI in cancer screening, focusing on breast and lung cancer.

Key Insights

Breast Cancer Screening

Accuracy and Performance:
- AI systems have shown generally good accuracy for breast cancer detection, with AUC values ranging from 69.2% to 97.8% .
- AI can outperform human radiologists in breast cancer prediction, reducing false positives and false negatives significantly .
Integration with Radiologists:
- Combining AI with radiologist expertise can improve screening outcomes, reducing the workload of radiologists by up to 88% while maintaining high accuracy .
- AI can be used to triage mammograms, potentially reducing radiologist workload by more than half and pre-emptively detecting a substantial proportion of cancers otherwise diagnosed later.
Methodological Concerns:
- Many studies use non-representative imaging data and have limited validation in external datasets, which raises concerns about the generalizability of AI models to real-world settings .
Primary Care Provider Preferences:
- Sensitivity is the most important attribute for primary care providers when considering AI for breast cancer screening. Most providers also prefer AI systems that involve radiologists in the interpretation process.

Lung Cancer Screening

Accuracy and Performance:
- AI systems can accurately classify and characterize lung nodules, showing high precision in lung cancer screening .
- AI models have been reported to perform as well as or better than experienced radiologists in distinguishing benign from malignant lung nodules, improving diagnostic accuracy and reducing false-positive rates .
Integration with Radiologists:
- AI can reduce the labor work of radiologists in reading LDCT and CXR images, increasing the accuracy of nodule detection and classification .
Future Potential:
- AI has the potential to assist in various aspects of lung cancer care, from detection and diagnosis to treatment decision-making and prognosis prediction.

Conclusion

AI has demonstrated significant potential in enhancing cancer screening, particularly for breast and lung cancers. It can improve diagnostic accuracy, reduce false positives and negatives, and alleviate the workload of radiologists. However, methodological concerns and the need for further validation in real-world settings remain critical challenges. Integrating AI with human expertise appears to be the most promising approach, ensuring high sensitivity and specificity while maintaining the involvement of radiologists in the diagnostic process.

Use of AI in cancer screening

Summary

Introduction

Key Insights

Breast Cancer Screening

Lung Cancer Screening

Conclusion

Sources and full results

Artificial Intelligence (AI) for the early detection of breast cancer: a scoping review to assess AI’s potential in breast screening practice

AI models show generally good accuracy in breast cancer detection, but methodological concerns and evidence gaps need to be addressed before it can be implemented in large-scale population-based screening.

Artificial intelligence to identify genetic alterations in conventional histopathology

AI methods can predict specific genetic alterations in tumor tissues from conventional hematoxylin and eosin (H&E) tissue slides, potentially reducing the workload of genetic analyses and aiding in precision oncology.

International evaluation of an AI system for breast cancer screening

An artificial intelligence system outperforms human experts in detecting breast cancer from mammograms, reducing false positives and false negatives, and potentially improving screening efficiency.

Artificial Intelligence and its future potential in lung cancer screening

Artificial intelligence and radiology integration can improve lung cancer screening sensitivity, specificity, and accuracy, potentially reducing morbidity and mortality.

Artificial Intelligence Evaluation of 122 969 Mammography Examinations from a Population-based Screening Program.

The AI system in mammography screening effectively detects 86.8% of screen-detected cancers and 92 of 205 interval cancers, with less than 20% not selected at the three evaluated thresholds.

Application of Artificial Intelligence in Lung Cancer

AI has the potential to improve lung cancer detection, diagnosis, and treatment by reducing labor work, increasing accuracy, and aiding in decision-making and prognosis prediction.

Artificial Intelligence Tools for Refining Lung Cancer Screening

AI algorithms can improve lung cancer screening by enhancing diagnostic accuracy and decreasing false-positive rates in low-dose computer tomography (LDCT) images.

Artificial intelligence (AI) to enhance breast cancer screening: protocol for population-based cohort study of cancer detection

AI algorithms can potentially enhance breast cancer screening by detecting cancers and interval cancers without contributing to overdiagnosis in real-world population screening programs.

Artificial intelligence in breast cancer screening: primary care provider preferences.

Most primary care providers (PCPs) accept the use of AI in breast cancer screening, with sensitivity being the most important attribute, and most accept AI determining likely negative mammograms without radiologist confirmation.

Effect of artificial intelligence-based triaging of breast cancer screening mammograms on cancer detection and radiologist workload: a retrospective simulation study.

Using AI-based triaging of breast cancer screening mammograms could potentially reduce radiologist workload by more than half and pre-emptively detect a substantial proportion of cancers otherwise diagnosed later.

Try another search