Himanshu J. Madhu

Extraction of medically interpretable features for classification of malignancy in breast thermography

Thermography, with high-resolution cameras, is being re-investigated as a possible breast cancer screening imaging modality, as it does not have the harmful radiation effects of mammography. This paper focuses on automatic extraction of medically interpretable non-vascular thermal features. We design these features to differentiate malignancy from different non-malignancy conditions, including hormone sensitive tissues and certain benign conditions, which have an increased thermal response. These features increase the specificity for breast cancer screening, which had been a long known problem in thermographic screening, while retaining high sensitivity. These features are also agnostic to different cameras and resolutions (up to an extent). On a dataset of around 78 subjects with cancer and 187 subjects without cancer, that have some benign diseases and conditions with thermal responses, we are able to get around 99% specificity while having 100% sensitivity. This indicates a potential break-through in thermographic screening for breast cancer. This shows promise for undertaking a comparison to mammography with larger numbers of subjects with more data variations.Thermography, with high-resolution cameras, is being re-investigated as a possible breast cancer screening imaging modality, as it does not have the harmful radiation effects of mammography. This paper focuses on automatic extraction of medically interpretable non-vascular thermal features. We design these features to differentiate malignancy from different non-malignancy conditions, including hormone sensitive tissues and certain benign conditions, which have an increased thermal response. These features increase the specificity for breast cancer screening, which had been a long known problem in thermographic screening, while retaining high sensitivity. These features are also agnostic to different cameras and resolutions (up to an extent). On a dataset of around 78 subjects with cancer and 187 subjects without cancer, that have some benign diseases and conditions with thermal responses, we are able to get around 99% specificity while having 100% sensitivity. This indicates a potential break-through in thermographic screening for breast cancer. This shows promise for undertaking a comparison to mammography with larger numbers of subjects with more data variations.

Mauka-mauka: measuring and predicting opportunities for webcam-based heart rate sensing in workplace environment

Prolonged sitting and physical inactivity at workplace often lead to various health risks such as diabetes, heart attack, cancer etc. Many organizations are investing in wellness programs to ensure the well-being of their employees. Generally wearable devices are used in such wellness programs to detect health problems of employees, but studies have shown that wearables do not result in sustained adoption. Heart rate measurement has emerged as an effective tool to detect various ailments such as anxiety, stress, cardiovascular diseases etc. There are pre-existing techniques that use webcam feed to sense heart rate subject to some experimental constraints like stillness of face, light illumination etc. In this paper, we show that in-situ opportunities can be found and predicted for webcam based heart rate sensing in the workplace environment by analyzing data from unobtrusive sensors in a pervasive manner.

Automatic Determination of Hormone Receptor Status in Breast Cancer Using Thermography

Estrogren and progesterone hormone receptor status play a role in the treatment planning and prognosis of breast cancer. These are typically found after Immuno-Histo-Chemistry (IHC) analysis of the tumor tissues after surgery. Since breast cancer and hormone receptor status affect thermographic images, we attempt to estimate the hormone receptor status before surgery through non-invasive thermographic imaging. We automatically extract novel features from the thermographic images that would differentiate hormone receptor positive tumors from hormone receptor negative tumors, and classify them though machine learning. We obtained a good accuracy of 82 % and 79 % in classification of HR\(+\) and HR− tumors, respectively, on a dataset consisting of 56 subjects with breast cancer. This shows a novel application of automatic thermographic classification in breast cancer prognosis.

A study of the effect of subject motion to pulse rate estimation

We presented a systematic study of how subject head motion affects pulse rate estimation using photoplethysmography from the subjects face. We evaluated the performance at various steps in the process, including object tracking, skin blob detection, pulse signal extraction and pulse rate estimation. We demonstrated that the signal-to-noise ratio of the power spectrum is a good indicator of signal artifacts induced by subject motion, thus can be used as a quantitative metric in continuous pulse rate monitoring to reduce estimation errors.We presented a systematic study of how subject head motion affects pulse rate estimation using photoplethysmography from the subjects face. We evaluated the performance at various steps in the process, including object tracking, skin blob detection, pulse signal extraction and pulse rate estimation. We demonstrated that the signal-to-noise ratio of the power spectrum is a good indicator of signal artifacts induced by subject motion, thus can be used as a quantitative metric in continuous pulse rate monitoring to reduce estimation errors.

Evaluating and improving the robustness of a video-based PR/RR monitoring system for a clinical environment

With the advancement and use of low cost cameras, video-based remote vital-monitoring technologies have drawn more and more attention. However, many of the previously proposed techniques rely on the assumption that the video is acquired under well-controlled conditions. There has been no study reported around how various acquisition factors and/or noise will affect the estimation accuracy. In this paper, we describe a systematic approach in understanding the challenges of extending the technology beyond a well-controlled environment and share the solution paths in making the technology robust in one specific application. The systematic approach involves first understanding the data analysis process, identifying key factors that affect the algorithms and then estimating the bounds of the factors and metrics that explain the variations in these vitals. As part of the solution paths, our paper goes into the details of how the algorithms were improved to address the findings of the current state. The refinement in the algorithms includes getting stronger signals in less controlled setting.

Marketing image categorization using hybrid human-machine combinations

Marketing instruments with nested, short-form, symbol loaded content need to be studied differently. Image classification in the Web2.0 world can dynamically use a configurable amount of internal and external data as well as varying levels of crowd-sourcing. Our work is one such examination of how to construct a hybrid technique involving learning and crowd-sourcing. Through a parameter called turkmix and a multitude of crowd-sourcing techniques available we show that we can control the trend of metrics such as precision and recall on the hybrid categorizer.