Multicentric study to evaluate the effectiveness of Thermalytix as compared with standard screening modalities in subjects who show possible symptoms of suspected breast cancer

Abstract

IntroductionMachine learning in computer-assisted diagnostics improves sensitivity of image analysis and reduces time and effort for interpretation. Compared to standard mammograms, a thermal scan is easily scalable and is a safer screening tool. We evaluate the performance of Thermalytix (an automated thermographic screening algorithm) compared with other standard breast cancer screening modalities.MethodsA prospective multicentre study was conducted to assess the non-inferiority of sensitivity of Thermalytix (test device) to that of standard modalities in detecting malignancy in subjects who show possible symptoms of suspected breast cancer. Standard screening modalities and Thermalytix were obtained and interpreted independently in a blinded fashion. A receiver operating characteristic (ROC) curve was constructed to identify the best cut-off point, non-inferiority margin of ≥10% to demonstrate the non-inferiority.ResultsWe recruited 258 symptomatic women who first underwent a thermal scan, followed by mammogram and/or ultrasound. At Youden’s Index of ROC curve, the test device had a sensitivity of 82.5% (95% CI 73.2 to 91.9) and specificity of 80.5% (95% CI 75.0 to 86.1) as compared with diagnostic mammogram, which had sensitivity of 92% (95% CI 80.7 to 97.8) and specificity of 45.9% (95% CI 34.3 to 57.9) when BI-RADS 3 (Breast Imaging-Reporting and Data System) was considered as test-positive. The overall area under the curve (AUC) was 0.845. For women aged <45 years, the test device had a sensitivity and specificity of 87.0% (95% CI 66.4 to 97.2) and 80.6% (95% CI 72.9 to 86.9), respectively. For women aged ≥45 years, the sensitivity and specificity were 80.5% (95% CI 65.1 to 91.2) and 86.5% (95% CI 78.0 to 92.6, respectively).ConclusionWe evaluated Thermalytix, a new AI-based modality for detecting breast cancer. The high AUC in both women under 45 years and above 45 years shows the potential of Thermalytix to be a supplemental diagnostic modality for all ages. Further evaluation on larger sample size is needed.Trial registration numberCTRI/2017/10/0 10 115;

Construction of machine learning-based models for cancer outcomes in low and lower-middle income countries: A scoping review

“…Further, only 6 models (4.4%) constructed for cancer outcomes in LLMIC populations have been externally validated (phase III implementation) all of which were within the last five years. These included models for breast cancer diagnosis ( 85 , 86 , 152 ), lung cancer diagnosis ( 51 , 52 ), head and neck cancer diagnosis ( 124 ), breast cancer metastasis ( 47 , 56 , 128 ), liver cancer risk prediction ( 78 ), and treatment response in colorectal cancer ( 148 ). Of these models, only two ( 47 , 56 , 85 , 152 ) sufficiently fulfilled the TRIPOD criteria for external validation based on the sample size.…”

“…These included models for breast cancer diagnosis ( 85 , 86 , 152 ), lung cancer diagnosis ( 51 , 52 ), head and neck cancer diagnosis ( 124 ), breast cancer metastasis ( 47 , 56 , 128 ), liver cancer risk prediction ( 78 ), and treatment response in colorectal cancer ( 148 ). Of these models, only two ( 47 , 56 , 85 , 152 ) sufficiently fulfilled the TRIPOD criteria for external validation based on the sample size. Also, none of the deep learning or deep hybrid learning models found have been assessed using external validation.…”

A prospective evaluation of breast thermography enhanced by a novel machine learning technique for screening breast abnormalities in a general population of women presenting to a secondary care hospital

“…In a previous multisite observational study of 470 symptomatic and asymptomatic women [22], the Thermalytix obtained a sensitivity of 91.02% and specificity of 82.39% with an overall area under the curve (AUC) of 0.90. In another publication [23] on a prospective multicenter study of 258 symptomatic women, an earlier version of the Thermalytix had a sensitivity of 82.5% and specificity of 80.5% with respect to the diagnostic mammogram, which had a sensitivity of 92% and specificity of 45.9%. While all the other studies of Thermalytix systems included patients who presented for either mammography or ultrasonography, in the current study, mammography was the reference standard for all participants.…”

Acceptance of artificial intelligence (AI)-based screening for breast health in urban slums of central Karnataka, India – SWOC analysis

“…[ 14 ] Though thermography as a screening tool was introduced in the 19 th century, it was not widely accepted then as the manual interpretation of thermograms is highly complex and subjective. [ 15 ] With recent advancements in high-resolution thermal cameras, breast thermography is re-emerging as a screening method. When thermography is combined with artificial intelligence (AI) or machine learning for automated analysis, the results are quantitative and consistent.…”

“…When thermography is combined with artificial intelligence (AI) or machine learning for automated analysis, the results are quantitative and consistent. [ 14 15 16 ] In resource-constrained settings such as in low and middle-income countries (LMICs), where the radiologist to population ratio is as low as 1:100,000, technology-assisted solutions can be scalable and pave the way forward. AI-powered algorithms for screening breast health conditions could reduce the need for unnecessary referrals, increase continuity with patients and enhance mastery for primary care physicians.…”