Hyperspectral imaging for Early Non-Invasive Detection of Wound Development on the Diabetic Foot


Motivation

Diabetes mellitus affects 194 million people worldwide and is expected to increase in prevalence to 439 million by the year 2030. Foot ulceration is a major complications which occurs in as many as 15-25% of type 1 and 2 patients with diabetes mellitus over their lifetime. If left untreated, foot ulcers may become infected and develop deep tissue necrosis which may require amputation. It has also been estimated that foot ulcers precede roughly 85% of all lower extremity amputations in patients with diabetes mellitus and more than 88,000 lower limb amputations are performed annually on diabetic patients in the United States. The cost of foot disorder diagnosis and management is estimated to be over $6 billion dollars annually in the United States. Furthermore, it has been suggested that 40% to 85% of diabetic foot amputations can be avoided with early detection and preventive techniques such as offloading and improved hygiene. Automated early identification of tissue at risk of ulcerating may enable directed care thereby reducing the incidence of foot ulceration and amputation. This study proposes to use hyperspectral imaging as a screening tool for detecting forming diabetic foot ulcers before tissue damage becomes apparent to a care-giver in a clinical setting.


Experimental Apparatus and Clinical Trial

Tissue oximetry measurements were performed during several visits with hyperspectral imaging of the feet in type 1 and 2 diabetes mellitus subjects that were at risk for foot ulceration.

  • Enrolled 153 patients with Type 1 and 2 Diabetes Mellitus at risk of developing foot ulcers.
  • Institutional Review Board: Olive View-UCLA IRB #05H-609300
  • Excluded patients with heart failure, stroke, or neurological complications.
  • Followed patients over a 24-week period for 11 total visits.
  • After signing the informed consent form, hyperspectral imaging of the feet was performed at each visit and preulcer sites analyzed retrospectively.
  • Patients were followed for 18 months.
  • The clinical study was performed by Dr. Aksone Nouvong, DPM

Oximetry Based Algorithm and Hypothesis

The data was retrospectively analyzed at 21 sites that ulcerate during the course of the study and an ulceration prediction index was developed. We calculated the maximum absolute difference (MAD) in OXY and DEOXY hemoglobin between sites that ulcerated and adjacent sites. The physiological premise of calculating MAD values relied on detecting hyperperfusion or hypovascularization associated with inflammation and ischemia in the target region, respectively. It was hypothesized that, ulceration due to ischemia may result in hypovascular tissue and thus, diminished oxyhemoglobin and deoxyhemoglobin concentrations in the affected site when compared with nearby unaffected tissue. On the other hand, ulceration accompanied by inflammation may result in hyperperfusion and thus, larger oxyhemoglobin and deoxyhemoglobin concentrations in the affected site when compared with nearby unaffected tissue.  Thus, we hypothesized that the absolute values |MAD(OXY)| and |MAD(DEOXY)| calculated at an ischemic or inflamed site should be larger than values calculated elsewhere on the foot or at locations on feet that did not form ulcers.

Then, an image processing algorithm based on this index was developed that could detect tissue at risk of ulceration.


Results

Scatter plot showing values of MAD(OXY) and MAD(DEOXY) determined for 21 affected and 21 contralateral sites from diabetic subjects who developed foot ulcers and from 100 random sites from diabetic subjects in the comparative groups. Values from affected areas were found to fall inside the grey regions where |MAD(OXY)|> 18 and |MAD(DEOXY)|> 5.8 with a p-value of less than 0.0001.

.

Composite image of three diabetic feet where the red overlay indicates that the maximum absolute differences in oxyhemoglobin and deoxyhemoglobin are such that |MAD(OXY)|> 18 and |MAD(DEOXY)|> 5.8. The approximate location of subsequent ulceration is circled.


Discussion

This algorithm was able to predict ulceration of tissue with a sensitivity and specificity of 95 and 80%, respectively, for images taken an average of 58 days before tissue damage was apparent to the naked eye. Receiver Operating Characteristic analysis was also performed to give a range of sensitivity/specificity values resulting in a Q-value of 89%.


Conclusion

This study demonstrated the ability of hyperspectral imaging between 500 and 650 nm to assess the risk of diabetic foot ulcer development. It also established that hyperspectral tissue oximetry has the ability to identify ischemic and inflammatory complications before they are visible to during a clinical examination. Retrospective analysis of hyperspectral tissue oximetry from preulcerative locations showed that diabetic foot ulcer formation can be predicted with a sensitivity and specificity of 95 and 80%, respectively. Thus, hyperspectral imaging could provide the diabetes care-giver with information necessary to treat and possibly prevent foot complications earlier and non-invasively. A prospective study combining the present algorithm with preventative care would prove the effectiveness of the present algorithm in reducing diabetic foot ulcer formation.


Publications

D. Yudovsky, A. Nouvong, K. Schomacker, and L. Pilon, 2011. Monitoring Temporal Development and Healing of Diabetic Foot Ulcer Using Hyperspectral Imaging, Journal of Biophotonics, Vol. 4, No. 7–8, pp. 565–576. doi: 10.1002/jbio.201000117.

D. Yudovsky and L. Pilon, 2011. Retrieving Skin Properties From In Vivo Diffuse Reflectance Measurements on Human Skin, Journal of Biophotonics, Vol. 4, No.5, pp.305-314, doi: 10.1002/jbio.201000069.

D. Yudovsky, A. Nouvong, K. Schomacker, and L. Pilon, 2011. Assessing Diabetic Foot Ulcer Development Risk with Hyperspectral Tissue Oximetry, Journal of Biomedical Optics, Vol.16, No.2, 026009. doi: 10.1117/1.3535592.

D. Yudovsky and L. Pilon, 2010. Modeling of Local Excitation Fluence Rate and Florescence Emission in Absorbing and Strongly Scattering Multilayered Media. Applied Optics, Vol. 49, No. 31, pp. 6072–6084. doi: 10.1364/AO.49.006072

D. Yudovsky , A. Nouvong, and L. Pilon, 2010. Hyperspectral Imaging for Diabetic Foot Wound Care, Journal of Diabetes Science and Technology, Vol.4, No.5, pp. 1099-1113.


Press Release

Diabetes foot ulcer diagnosis and treatment could be improved and made earlier with hyperspectral imaging techniques by Matthew Chin