Tissue viability by multispectral near infrared imaging: A fuzzy C-means clustering analysis

Clinically, skin color, temperature, and capillary perfusion are used to as sess tissue viability following microvascular tissue transfer, However, cli nical signs that arise as a consequence of poor perfusion become evident on ly after several hours of compromised perfusion. This study demonstrates th e potential usefulness of optical/infrared multispectral imaging in the pro gnosis of tissue viability immediately post-surgery. Multispectral images o f a skin flap model acquired within 1 h of surgical elevation are analyzed in comparison to the final 72-h clinical outcome with a high degree of corr elation, Regional changes in tissue perfusion and oxygenation present immed iately following surgery are differentiated using fuzzy clustering and imag e processing algorithms, These methodologies reduce the intersubject variab ility inherent in infrared imaging methods such that the changes in perfusi on are reproducible and clearly distinguishable across all subjects. Clinic ally, an early prognostic indicator of viability such as this would allow f or a more timely intervention following surgery in the event of compromised microvasculature.