Back

Point-of-care thermal imaging tool assisting in the development of imaging biomarker as pre-diagnostics of graves' disease

Goswami, M.; Dhingra, V. K.; Kumar, A.; Sharma, V.

2022-07-07 radiology and imaging
10.1101/2022.07.05.22277045 medRxiv
Show abstract

This article presents the development of a tool to address the scarcity of clinically validated datasets for Thyroid related abnormalities. Infrared thermography Images are taken from each of the sixty persons of different age groups and gender. Scintigraphy and standard thyroid blood test results are used to categorize these persons into thirty-three females and thirteen males suffering from graves disease. Eleven females and three males are found to be in healthy conditions and used as control. An Artificial Intelligence algorithm is used to automatically segment and extract the histogram-associated information within the thyroid and cheek region from the collected images. A very simple and novel imaging biomarker is found to be moderately correlated w.r.t age and gender. A smartphone app integrated with a dedicated smartphone-based compact IR camera add-on is developed and deployed in a clinical environment to enrich the analysis. This point-of-care tool is expected to categorize healthy cases from patients automatically. It is to reduce the ethical burden on clinicians shoulders before recommending radioactive contamination-prone Scintigraphy and/or expensive and relatively slower thyroid blood tests. If adopted, such preliminary tests will save costs on patients end and burden at pathology labs, especially in densely populated countries such as India.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.