Diabetic foot is one of the major complications experienced by diabetic patients. Thus, early detection and appropriate treatment can prevent traumatic outcomes such as lower limb amputation. The temperature variations on the plantar foot can be related to diabetic foot complications. Infrared thermography allows qualitative and visual documentation of temperature fluctuation in vascular tissues. This work introduces a computer aided detection (CAD) system for diabetic foot using plantar foot thermograms. The proposed system implements image processing methods that yielded 89.39% accuracy using only five features.
The proposed technology has the following stages:
The plantar foot thermograms are pre-processed to segment the plantar foot regions and then warped into uniform size.
2. Image decomposition
The warped foot images are decomposed using Discrete Wavelet Transform (DWT) and Higher Order Spectra (HOS).
3. Feature extraction
Entropies and texture features are extracted from the decomposed bilateral foot images.
4. Feature selection
Student t-test is applied to select and rank the significant features.
Support Vector Machine (SVM) classifier is used for classification.
The thermograms acquired are pre-processed by segmenting the plantar foot regions using polygon and then proceed to warp all the segmented plantar foot regions into uniform size. Afterward, the warped grayscale foot images are decomposed using Discrete Wavelet Transform (DWT) and Higher Order Spectra (HOS) prior to extracting texture and entropy features. The features values from left and right foot are subtracted. Subsequently, student t-test is applied on the resultant features to select and rank the significant features. Lastly, the 27 significantly ranked features (p value
This technology will be applicable to all healthcare industry related to detection of diabetic foot. The proposed technology can be introduced as an adjunctive diabetic foot screening tool in clinics to effectively assist podiatrist in decision making and diagnosis processes. Further, such methods can be extended to other diseases like cancers, wound healing, eye diseases, and bone fractures etc.
The technique is fast, non-invasive and non-contact.
It can be easily programmed and installed in any clinician’s laboratory.
The expert training is not preferred as results obtained are highly objective.
No inter-observer variability and reproducible as compared to manual diagnosis.
The detection support system helps in reducing the workload of healthcare professionals by providing real time medical support in diagnosing the magnitude of diabetic foot complications.