Key points are not available for this paper at this time.
Abstract Background Breast cancer is the leading cause of cancer-related deaths among women. Early detection is crucial for improving treatment outcomes and reducing costs. Systematic screening programs using mammography pose significant challenges in developing countries due to high-costs and skill shortages. Thermalytix is an affordable, portable, artificial intelligence (AI) based test that has demonstrated good clinical efficacy and economic feasibility for population-screening. This paper presents insights and data from implementing Thermalytix test on over 100,000 women in India. Methods Thermalytix was deployed at 150 clinical sites and at 1000 + screening camps outside hospitals. All women who took the test with informed consent, in either of these modes, were included to form a diverse cohort of 104,411 women from various socioeconomic backgrounds across 20 Indian-states. Thermalytix AI algorithms analyzed thermal patterns and automatically triaged women into three risk categories (red-yellow-green). Test Positivity Rate (TPR), assuming Red as test-positive, was computed for different cohorts. Results Thermalytix showed a TPR of 6.64% across the entire population. TPR in symptomatic women was 4x higher than in asymptomatic women. Women tested in hospitals exhibited a 1.6x higher TPR than those tested in screening camps. Highest TPR was observed in women aged above 60, followed by those aged 41–50 with complaints and those aged 31–40 without complaints. Postmenopausal women had a higher TPR than premenopausal women. Prior breast cancer led to a higher TPR than those without. Conclusion This study demonstrated the feasibility of implementing Thermalytix for community screening in resource-constrained countries, and the findings correlated with known risk-factors.
Kakileti et al. (Thu,) studied this question.