Hyperspectral imaging for skin assessment in systemic sclerosis: a pilot study.

Area of Science:

• Medical Imaging
• Dermatology
• Rheumatology

Background:

• Systemic sclerosis (SSc) skin tightness severity correlates with internal organ involvement.
• Current clinical assessment of SSc skin tightness, like the modified Rodnan skin score, lacks standardization and shows high variability.
• There is a need for objective, standardized methods to assess SSc skin changes.

Purpose of the Study:

• To compare the efficacy of hyperspectral imaging (HSI) against traditional methods like skin scores and ultrasound in differentiating SSc patients from healthy controls.
• To evaluate HSI's ability to detect subclinical skin changes in SSc patients.

Main Methods:

• Utilized short-wave infrared light for hyperspectral imaging (HSI) to calculate the spectral angle mapper (SAM).
• Assessed skin severity using skin scores, ultrasound (dermal thickness), and strain elastography.
• Employed Spearman's correlation to analyze relationships between skin scores, strain ratio, thickness, and SAM.
• Used receiver operating characteristic (ROC) curves to compare the diagnostic performance of different assessment tools.

Main Results:

• Spectral angle mapper (SAM) positively correlated with skin scores and dermal thickness in SSc patients.
• SAM values were significantly higher in SSc patients, even those with normal skin scores, compared to healthy controls.
• HSI, measured by SAM, demonstrated the highest diagnostic accuracy (AUC: 0.812) for detecting SSc compared to skin scores (AUC: 0.712), ultrasound thickness (AUC: 0.585), and elastography strain ratio (AUC: 0.522).

Conclusions:

• Spectral angle mapper (SAM) derived from HSI is correlated with clinical skin scores and sensitive enough to detect subclinical SSc.
• Hyperspectral imaging (HSI) offers a novel, non-invasive approach for quantitatively assessing skin alterations in systemic sclerosis.
• HSI shows potential as a standardized tool for monitoring SSc progression and treatment response.