Trial Outcomes & Findings for A Prospective Study to Evaluate Clinical Performance of Thermalytix in Detecting Breast Cancers (NCT NCT04688086)

NCT ID: NCT04688086

Last Updated: 2024-10-17

Results Overview

To assess the clinical performance of Thermalytix as compared to standard screening modalities. Sensitivity, specificity, positive predictive value and negative predictive value of Thermalytix

• Recruitment status: COMPLETED
• Target enrollment: 687 participants
• Primary outcome timeframe: 2 days
• Results posted on: 2024-10-17

Participant Flow

Participant milestones

Measure	Women With no Personal History of Breast Cancer Women who came in for a breast mammography between ages 30 and 80 years were invited to take part in the study. All the women included in the study underwent breast cancer screening first by Thermalytix, the AI-based thermal imaging test, followed by mammography. Thermalytix: Thermalytix is an Artificial intelligence based automated breast screening solution that analyzes thermal distribution on the breast to generate a breast health score automatically. Thermal imaging was performed by a trained technician to capture thermal images of the participant in five views. These thermal images were uploaded to Thermalytix software on the cloud where it was automatically analyzed by AI-based Thermalytix computer-aided detection (CADe) engine. This CADe engine analyzes uploaded thermal images and outputs an interpretation report for each participant with quantitative scores corresponding to computed probability of malignancy based on the structural, vascular, areolar, thermal properties of the observed abnormality. Thermalytix also generates annotated images with markings of abnormal regions and an overall Thermalytix score suggesting likelihood of breast malignancy. The locked AI model Thermalytix algorithm version 3, dated December 2018 was used for the analysis.
Overall Study STARTED	687
Overall Study COMPLETED	459
Overall Study NOT COMPLETED	228

Reasons for withdrawal

Measure	Women With no Personal History of Breast Cancer Women who came in for a breast mammography between ages 30 and 80 years were invited to take part in the study. All the women included in the study underwent breast cancer screening first by Thermalytix, the AI-based thermal imaging test, followed by mammography. Thermalytix: Thermalytix is an Artificial intelligence based automated breast screening solution that analyzes thermal distribution on the breast to generate a breast health score automatically. Thermal imaging was performed by a trained technician to capture thermal images of the participant in five views. These thermal images were uploaded to Thermalytix software on the cloud where it was automatically analyzed by AI-based Thermalytix computer-aided detection (CADe) engine. This CADe engine analyzes uploaded thermal images and outputs an interpretation report for each participant with quantitative scores corresponding to computed probability of malignancy based on the structural, vascular, areolar, thermal properties of the observed abnormality. Thermalytix also generates annotated images with markings of abnormal regions and an overall Thermalytix score suggesting likelihood of breast malignancy. The locked AI model Thermalytix algorithm version 3, dated December 2018 was used for the analysis.
Overall Study Lost to Follow-up	164
Overall Study Imaging error;	64

Baseline Characteristics

Race and Ethnicity were not collected from any participant.

Baseline characteristics by cohort

Measure	Women With no Personal History of Breast Cancer n=459 Participants Women who came in for a breast mammography between ages 30 and 80 years were invited to take part in the study. All the women included in the study underwent breast cancer screening first by Thermalytix, the AI-based thermal imaging test, followed by mammography. Thermalytix: Thermalytix is an Artificial intelligence based automated breast screening solution that analyzes thermal distribution on the breast to generate a breast health score automatically. Thermal imaging was performed by a trained technician to capture thermal images of the participant in five views. These thermal images were uploaded to Thermalytix software on the cloud where it was automatically analyzed by AI-based Thermalytix computer-aided detection (CADe) engine. This CADe engine analyzes uploaded thermal images and outputs an interpretation report for each participant with quantitative scores corresponding to computed probability of malignancy based on the structural, vascular, areolar, thermal properties of the observed abnormality. Thermalytix also generates annotated images with markings of abnormal regions and an overall Thermalytix score suggesting likelihood of breast malignancy. The locked AI model Thermalytix algorithm version 3, dated December 2018 was used for the analysis.
Age, Categorical <=18 years	0 Participants n=459 Participants
Age, Categorical Between 18 and 65 years	441 Participants n=459 Participants
Age, Categorical >=65 years	18 Participants n=459 Participants
Sex: Female, Male Female	459 Participants n=459 Participants
Sex: Female, Male Male	0 Participants n=459 Participants
Region of Enrollment India	459 participants n=459 Participants

PRIMARY outcome

Timeframe: 2 days

Population: All women who satisfied the inclusion criteria

To assess the clinical performance of Thermalytix as compared to standard screening modalities.

Sensitivity, specificity, positive predictive value and negative predictive value of Thermalytix

Outcome measures

Measure	Women With no Personal History of Breast Cancer n=459 Participants Women who came in for a breast mammography between ages 30 and 80 years were invited to take part in the study. All the women included in the study underwent breast cancer screening first by Thermalytix, the AI-based thermal imaging test, followed by mammography. Thermalytix: Thermalytix is an Artificial intelligence based automated breast screening solution that analyzes thermal distribution on the breast to generate a breast health score automatically. Thermal imaging was performed by a trained technician to capture thermal images of the participant in five views. These thermal images were uploaded to Thermalytix software on the cloud where it was automatically analyzed by AI-based Thermalytix computer-aided detection (CADe) engine. This CADe engine analyzes uploaded thermal images and outputs an interpretation report for each participant with quantitative scores corresponding to computed probability of malignancy based on the structural, vascular, areolar, thermal properties of the observed abnormality. Thermalytix also generates annotated images with markings of abnormal regions and an overall Thermalytix score suggesting likelihood of breast malignancy. The locked AI model Thermalytix algorithm version 3, dated December 2018 was used for the analysis.
Sensitivity, Specificity, Positive Predictive Value and Negative Predictive Value of Thermalytix Sensitivity	95.24 percentage Interval 76.18 to 99.88
Sensitivity, Specificity, Positive Predictive Value and Negative Predictive Value of Thermalytix Specificity	88.58 percentage Interval 85.23 to 91.41
Sensitivity, Specificity, Positive Predictive Value and Negative Predictive Value of Thermalytix PPV	28.57 percentage Interval 18.4 to 40.62
Sensitivity, Specificity, Positive Predictive Value and Negative Predictive Value of Thermalytix NPV	99.74 percentage Interval 98.58 to 99.99

SECONDARY outcome

Timeframe: 2 days

Population: Women with breast density ACR category C or D

To assess the influence of patient characteristics (Breast density) on the diagnostic accuracy of Thermalytix.

Outcome measures

Measure	Women With no Personal History of Breast Cancer n=168 Participants Women who came in for a breast mammography between ages 30 and 80 years were invited to take part in the study. All the women included in the study underwent breast cancer screening first by Thermalytix, the AI-based thermal imaging test, followed by mammography. Thermalytix: Thermalytix is an Artificial intelligence based automated breast screening solution that analyzes thermal distribution on the breast to generate a breast health score automatically. Thermal imaging was performed by a trained technician to capture thermal images of the participant in five views. These thermal images were uploaded to Thermalytix software on the cloud where it was automatically analyzed by AI-based Thermalytix computer-aided detection (CADe) engine. This CADe engine analyzes uploaded thermal images and outputs an interpretation report for each participant with quantitative scores corresponding to computed probability of malignancy based on the structural, vascular, areolar, thermal properties of the observed abnormality. Thermalytix also generates annotated images with markings of abnormal regions and an overall Thermalytix score suggesting likelihood of breast malignancy. The locked AI model Thermalytix algorithm version 3, dated December 2018 was used for the analysis.
Sensitivity and Specificity of Thermalytix for Women With Dense Breast Tissue (ACR Category C or D) Sensitivity	100 percentage Interval 69.15 to 100.0
Sensitivity and Specificity of Thermalytix for Women With Dense Breast Tissue (ACR Category C or D) Specificity	81.65 percentage Interval 74.72 to 87.35

SECONDARY outcome

Timeframe: 2 days

Population: Women with Breast Density ACR category "A" or "B"

To assess the influence of patient characteristics (Breast density) on the diagnostic accuracy of Thermalytix.

Sensitivity and Specificity of Thermalytix in Women with Fatty Breasts - Breast Density ACR category "A" or "B"

Outcome measures

Measure	Women With no Personal History of Breast Cancer n=291 Participants Women who came in for a breast mammography between ages 30 and 80 years were invited to take part in the study. All the women included in the study underwent breast cancer screening first by Thermalytix, the AI-based thermal imaging test, followed by mammography. Thermalytix: Thermalytix is an Artificial intelligence based automated breast screening solution that analyzes thermal distribution on the breast to generate a breast health score automatically. Thermal imaging was performed by a trained technician to capture thermal images of the participant in five views. These thermal images were uploaded to Thermalytix software on the cloud where it was automatically analyzed by AI-based Thermalytix computer-aided detection (CADe) engine. This CADe engine analyzes uploaded thermal images and outputs an interpretation report for each participant with quantitative scores corresponding to computed probability of malignancy based on the structural, vascular, areolar, thermal properties of the observed abnormality. Thermalytix also generates annotated images with markings of abnormal regions and an overall Thermalytix score suggesting likelihood of breast malignancy. The locked AI model Thermalytix algorithm version 3, dated December 2018 was used for the analysis.
Sensitivity and Specificity of Thermalytix in Women With Breast Density ACR Category "A" or "B" Sensitivity	90.9 percentage Interval 73.9 to 100.0
Sensitivity and Specificity of Thermalytix in Women With Breast Density ACR Category "A" or "B" Specificity	92.5 percentage Interval 89.4 to 95.6

Adverse Events

Women With no Personal History of Breast Cancer

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Adverse event data not reported

Additional Information

Dr Sathiakar Collison

NIRAMAI Health Analytix Private Limited

Phone: +919810823838

Email: sathiakar.collison@niramai.com

Results disclosure agreements

• Principal investigator is a sponsor employee
• Publication restrictions are in place