Clinical and Technical publications using Niramai Technology
November 2022

Please Note : The soft copies of the documents listed on this page may be subjected to copyright of a Journal or conference publisher. It has been shared here for academic use only. Please do not redistribute .

Please Note: Niramai holds 32 granted patents in the area of thermal imaging and AI. The technical content and ideas in the below publications are protected by multiple patents. Reuse of any idea is highly discouraged.

Clinical and Technical publications are listed in reverse chronological order below. First list includes full papers published in journals and second list include posters and abstract. All are peer reviewed publications.

Peer-Reviewed Full Paper Publications in Medical and AI journals:

  1. 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, Frontiers in Artificial Intelligence, Jan 2023 (online link)
    → A blinded prospective comparison of Thermalytix and Mammography on 459 women in Max Hospital Delhi.
  2. Feasibility and outcomes of using a Novel Artificial Intelligence Enhanced Breast Thermography technique, Thermalytix, in screening for breast abnormalities at primary health centres at the community level in South India” , International Journal Of Community Medicine And Public Health, Accepted Oct 2022 (online link)
    → An analysis of a total of 6935 women who underwent AI-powered Thermalytix test in BBMP referral hospitals.
  3. Automated vascular analysis of breast thermograms with interpretable features. Journal of Medical Imaging (Bellingham). 2022 July
    → This paper presents blood vessel segmentation and classification algorithms used in Thermalytix for generating breast cancer screening report with interpretable features to aid radiologists in understanding the predictions.
  4. Multicentric study to evaluate the effectiveness of Thermalytix as compared with standard screening modalities in subjects who show possible symptoms of suspected breast cancer. BMJ Open. 2021;11(10):e052098. Published Oct 2021, doi:10.1136/bmjopen-2021-052098
    → A multisite prospective study that evaluates the diagnostic accuracy of Thermalytix compared to standard imaging modalities. Evaluates an early version of Thermalytix algorithm (machine learning model dated 2017) on only symptomatic patients.
  5. A Systematic Study to Evaluate the Benefit of using a Visualization Tool for Breast Thermography,,  Thermology International, American Academy of Thermology, May 2021
    → This journal paper describes SMILE-100 a visualization tool with automated annotation of hotspots as an aid to thermographers.
  6. Observational Study to Evaluate the Clinical Efficacy of Thermalytix for Detecting Breast Cancer in Symptomatic and Asymptomatic Women’, Journal of Clinical Oncology Global Oncology (JCO GO) from American Society of Clinical Oncology (ASCO). Oct 2020; 6:1472-1480.
    → A systematic study on 470 participants evaluating clinical efficacy of Niramai’s Thermalytix solution (machine learning model dated 2018) for detecting breast cancer by comparing with current standard tests of mammography, breast ultrasound and biopsy.
  7. Personalized risk prediction for breast cancer pre-screening using artificial intelligence and thermal radiomics. Artificial Intelligence in Medicine. 2020; 105:101854, Elsevier.
    → Describes the technical algorithm used in Thermalytix with the evaluation results, in the setting of risk prediction.
  8. Estimation of Breast Cancer Incidence Risk in Presence of Incomplete or Inaccurate Information. Asian Pacific Journal of Cancer Prevention. 2020;21(8):2307-2313.
    →Describes a new deep learning model which computes 1 year breast cancer risk using data from a questionnaire of non-imaging data (risk prediction).
  9. Cascaded CNN for View Independent Breast Segmentation in Thermal Images. Annu Int Conf IEEE EMBC Engineering in Medicine and Biology Society. 2019;
    →Describes the deep learning model used in Thermalytix for segmenting region of interest in breast thermal images
  10. Extraction of medically interpretable features for classification of malignancy in breast thermography. Annu Int Conf IEEE Engineering in Medicine and Biology Society. 2016;1062-1065.
    → First version of the Thermalytix algorithm described with focus on semantic features used for the machine learning classifier.
  11. Automated blood vessel extraction in two-dimensional breast thermography. In 2016 IEEE International Conference on Image Processing (ICIP) 2016 Sep 25 (pp. 380-384). IEEE. 
    → Describes the algorithm used in Thermalytix for extracting blood vessel structure from breast thermal images.
  12. Automatic determination of hormone receptor status in breast cancer using thermography. In International Conference on Medical Image Computing and Computer-Assisted Intervention 2016 Oct 17 (pp. 636-643). Springer, Cham.
    → Describes a new machine learning algorithm to determine hormonal status of breast cancer using just breast thermal images.
  13. Exploring deep learning networks for tumour segmentation in infrared images. Quantitative InfraRed Thermography Journal. 2020 Jul 2;17(3):153-68.
    → Describes a deep learning approach to determine hotspot in breast thermal images that may represent potential tumours.
  14. Robust Estimation of Breast Cancer Incidence Risk in Presence of Incomplete or Inaccurate Information. Asian Pacific Journal of Cancer Prevention 21, no. 8 (2020): 2307-2313, 2020
    → A machine learning based approach for risk assessment without thermal images
  15. Thermalytixusing AI to save lives, XRDS: Crossroads, The ACM Magazine for Students 26, no. 3 (2020): 38-41
    → A simple explanation of NIRMAI Thermalytix in this student facing magazine from ACM
  16. “Semi-automated breast cancer tumor detection with thermographic video imaging,” IEEE Int. Conf. Engineering, Medicine and Biology Society, August 2015, Milan,
    → An early version of the algorithm which used simple image processing techniques.
  17. “Initial evaluation of human supervised automated breast cancer screening using thermography,” Quantitative InfraRed Thermography Asia Conference, July 6-10, 2015, Mamallapuram. (PDF )
    → Our first lab level evaluation of machine learning algorithm on thermal images
  18. “Method for classifying cancerous and normal regions in breast thermography for small size tumors,” Quantitative InfraRed Thermography Asia Conference, July 6-10, 2015, Mamallapuram
    → A very early version of the classification technique
  19. Book Chapter: Advances in breast thermography. In New Perspectives in Breast Imaging, InTech. 2017:91. doi:10.5772/intechopen.69198.
  20. Book Chapter: AIM for Breast Thermography, Artificial Intelligence in Medicine, 2021

    Presentations in International Conferences published as abstracts or posters

  21. Analyzing the Performance of Thermalytix, an AI-based Breast Cancer Screening Test in a Community Setting, Geetha Manjunath, Charitha Gangadharan, Sathiakar Collison, Lakshmi K, Gargi D, Purnima Madhivanan, Karl Francis Krupp, San Antonio Breast Cancer Symposium, SABCS, Texas, Dec 2022
    →This study analyses 6935 women screened with Thermalytix in 22 BBMP hospitals
  22. An Automated Risk Stratification System for Breast Cancer Screening using Thermalytix, Siva Teja Kakileti, Himanshu J Madhu, Richa Bansal, Akshita Singh, Sudhakar Sampangi, Bharat Aggarwal, Geetha Manjunath, San Antonio Breast Cancer Symposium, SABCS, Dec 2022
    → This analysis the use of B-Score (generated by Thermalytix) for risk stratified screening
  23. Real world evidence from a Breast Screening pilot for the underprivileged : Experiences from Bruhat Bengaluru Mahanagara Palike hospitals, European Breast Cancer Conference, Nov 2022, Barcelona, Spain (
    →This study analyses 6935 women screened with Thermalytix in 22 BBMP hospitals
  24. MaThAI : A MultiModal imaging combining Mammography and Thermalytix for better prioritization of mammography scans to detect early malignancies, European Breast Cancer Conference, Nov 2022, Barcelona, Spain (
    → In this paper, we propose and evaluate a  multimodal imaging modality called MaThAI that combines mammography and Thermalytix for prioritization of Mammography scans using Thermalytix B-Score.
  25. Evaluating screening performance of artificial intelligence-based Thermalytix by comparing breast lesion sizes detected by Thermalytix with mammography, UK Imaging and Oncology, UKIO 2022, Liverpool UK. (online link awaited)
    → In this posthoc analysis of data from a study on 470 subjects, we show that Thermalytix had the ability to detect even small lesions and correlated with mammographyto detect even small lesions and correlated with mammography.
  26. Performance of artificial intelligence-based breast cancer screening in a community setting: a real-world evaluation study. The Lancet Oncology. 2022 Jul 1;23:S20. , in a community setting, The Lancet Oncology, ISSN: 1470-2045, Vol: 23, Page: S20, Published 2022 (online link).
    → A funnel analysis of results on 13,933 women  screened with Thermalytix in community setting. (corresponding full paper under preparation)
  27. AI for early-stage breast cancer screening. ASCO Breakthrough Summit, Bangkok, Thailand, Oct 2019. Adjudged as top 50 presentations.
    → An observational study on 769 participants evaluating clinical efficacy of Niramai’s Thermalytix solution for detecting breast cancer by comparing with current standard tests of mammography, breast ultrasound and biopsy
  28. A prospective study of an AI-based breast cancer screening solution for resource- constrained settings. Abstract in ASCO Annual Meeting 2021.
    → A systematic A study comparing Thermalytix and mammography on 459 participants (corresponding detailed journal manuscript is under review)
  29. Prospective evaluation of an affordable and portable artificial intelligence-based modality for breast cancer screening in women across breast densities, Abstract in 2022 ASCO Annual Meeting,
    → Analysis of the results from the prospective study with additional emphasis on results on dense breasts
  30. Thermal Radiomics For Detection Of Axillary Lymph Node Metastases A Pilot Study, 10th Global Breast Cancer Conference (GBCC 10), April 8 – 10, 2021 I Grand Walkerhill Seoul, Korea page 51 PO138
    → A pilot study investigating the strength of the Thermalytix test to detect axilla/lymph node metastatis in breast cancer patients.
  31. Correlation of Thermalytix – an artificial intelligence based thermal breast screening tool in detecting a breast lesion as benign, malignant or normal, SABCS San Antonio Breast Cancer Symposium International Conference, San Antonio, Texas, USA, Dec 2020 (PS2-44).
    → A pilot study comparing the B-Score provided by Thermalytix with BI-RADS scoring provided by standard imaging tests and histopathological results.
  32. A Non-Radiation Based Screening to Detect Dense Breasts. SABCS San Antonio Breast Cancer Symposium International Conference, San Antonio, Texas, USA, Dec 2019.
    → An evaluation of Thermalytix algorithms for detecting breast density class using breast thermal images alone.
  33. Artificial Intelligence over thermal images for radiation-free breast cancer screening. SABCS San Antonio Breast Cancer Symposium International Conference on Breast Cancer Research, Dec 2018. ,
    → First poster comparing Thermalytix with mammography on 247 participants showed the benefit of using Thermalytix on BIRAD 0 mammograms.
  34. Machine Learning over Thermal Images for Accurate Breast Cancer Screening, International Conference on Advances in Breast Cancer Treatments, Kyoto, Japan, April 26-27, 2018.
    → A systematic study evaluating the performance of AI-based Thermalytix over manual thermography (visual analysis of thermal images).
  35. Thermalytix: An Advanced Artificial Intelligence Based Solution for Non-Contact Breast Screening. International Conference on Breast Cancer Management, Amsterdam, February, 12-13, 2018.
    → First pilot study comparing Thermalytix with mammography on 147 participants.
  36. Evaluation of efficacy of thermographic breast imaging in breast cancer : A Pilot Study, Breast Disease 2016, 36(4), 143-147,