Automated Prediction of Diabetes Mellitus using Infrared Thermal Foot Images: Recurrent Neural Network Approach.

Millions of people worldwide are affected by Diabetes Mellitus (DM), which is a chronic disease. Evaluation of the DM indicator, namely blood glucose level, requires invasive methods such as glucometer or blood tests, which cause discomfort to the patient. Automated noninvasive monitoring methods are urgently needed to ensure consistency and better treatment. The regular monitoring of DM can prevent or delay the onset of complications. Thermal foot images have been proposed as noninvasive methods for the prediction of DM. Thermograms were acquired at Mittal Eye Hospital, Sangrur, India, from 50 participants in the diabetic (without neuropathic conditions) and non-diabetic groups using a thermal camera (FLIR E-60). This study proposes an automated prediction system for DM using thermal foot images and Recurrent Neural Network (RNN) approach. The proposed system processes the thermal images and extracts relevant features using a CNN (Convolutional Neural Network). The extracted features were then fed to the RNN to predict the presence or absence of the DM. The experimental results demonstrate that the proposed framework attains an accuracy of (97.14 ± 1.5) %, surpassing the predictive capabilities of light-weight convolutional neural network (Lw-CNN), which only achieves an accuracy of (82.9 ± 3) % in predicting DM. This performance outperformed other state-of-the-art methods in the field. Our approach has the potential to be used as prediction tool for DM. Therefore, the proposed system has the potential for prediction of DM and improve patient outcomes by enabling timely intervention. Future work should focus on evaluating the proposed system on a larger dataset and integrating it with clinical decision support systems for personalized care. This study holds the promise of transforming DM screening and diagnosis, leading to enhanced patient outcomes.

Thermography as an Economical Alternative Modality to Mammography for Early Detection of Breast Cancer.

Breast cancer has become a menacing form of cancer among women accounting for 11.6% of total deaths of 9.6 million due to all types of cancer every year all over the world. Early detection increases chances of survival and reduces the cost of treatment as well. Screening modalities such as mammography or thermography are used to detect cancer early; thus, several lives can be saved with timely treatment. But, there are interpretational failures on the part of the radiologists to read the mammograms or thermograms and also there are interobservational and intraobservational differences between them. So, the degree of variations among the different radiologists in the interpretation of results is very high resulting in false positives and false negatives. The double reading can reduce the human errors involved in the interpretation of mammograms. But, the limited number of medical professionals in developing or underdeveloped countries puts a limitation on this remedial way. So, a computer-aided system (CAD) is proposed to detect the benign cases from the abnormal cases that can result in automatic detection of breast cancer or can provide a double reading in the case of nonavailability of the trained medical professionals in developing economies. The generally accepted screening modality is mammography for the early detection of cancer. But thermography has been tried for early detection of breast cancer in recent times. The high metabolic activity of the cancer cells results in an early change in the temperature profile of the region. This shows asymmetry between normal and cancerous breast which can be detected using different techniques. Thus, this work is focussed on the use of thermography in the early detection of breast cancer. An experimental study is conducted to find the results of classification accuracy to compare the efficacy of thermography and mammography in classifying the normal from abnormal ones and further abnormal ones into benign and malignant cases. Thermography is found to have classification accuracy almost at par with mammography for classifying the cancerous breasts from healthy ones with classification accuracies of thermography and mammography being 96.57% and 98.11%, respectively. Thermography is found to have much better accuracy in identifying benign cases from the malignant ones with the classification accuracy of 92.70% as compared to 82.05% with mammography. This will result in the early detection of cancer. The advantage of being portable and inexpensive makes thermography an attractive modality to be used in economically backward rural areas where mammography is not practically possible.

An automated approach to enhance the thermographic evaluation on orofacial regions in lateral facial thermograms.

Segmentation of characteristic facial regions has often been an initial step of thermographic analysis in face recognition and clinical diagnosis. Moreover, fast and accurate thermographic analysis on characteristic areas is highly reliant on segmentation approach. Usually, frontal and lateral projections are used to capture the facial thermograms. The significant role of lateral facial thermography to diagnose the various problems associated with orofacial regions has been remarkable in many studies. So far, no study has presented an automatic approach for the segmentation of characteristic areas in lateral facial thermograms. For this purpose, an automatic approach to segment the characteristic areas in lateral facial thermograms is proposed. The dataset of 140 facial thermograms with 1 left and 1 right lateral view per subject is created. Initially, image binarization is performed using optimal temperature thresholding for better visualization of facial geometry. Then, the automatic localization of characteristic points is performed at two levels, based on (a) geometrical features of the face, and (b) local thermal patterns. At last, the characteristic points and auxiliary points are used to segment the characteristic areas in the orofacial region of the face. To evaluate the segmentation performance of the proposed methodology, the automatically localized characteristic points are compared with manually marked using Euclidean distance based comparison criterion. With the localization error δch_pt≤0.05, the proposed automatic approach shows 92.04% of overall localization accuracy and 85% of overall segmentation accuracy. The key conclusion is that the proposed algorithm can potentially automate the process of thermographic analysis on characteristic areas to assist the diagnosis of problems in the orofacial region.

Comparative Analysis of Hybrid Models for Prediction of BP Reactivity to Crossed Legs.

Crossing the legs at the knees, during BP measurement, is one of the several physiological stimuli that considerably influence the accuracy of BP measurements. Therefore, it is paramount to develop an appropriate prediction model for interpreting influence of crossed legs on BP. This research work described the use of principal component analysis- (PCA-) fused forward stepwise regression (FSWR), artificial neural network (ANN), adaptive neuro fuzzy inference system (ANFIS), and least squares support vector machine (LS-SVM) models for prediction of BP reactivity to crossed legs among the normotensive and hypertensive participants. The evaluation of the performance of the proposed prediction models using appropriate statistical indices showed that the PCA-based LS-SVM (PCA-LS-SVM) model has the highest prediction accuracy with coefficient of determination (R2) = 93.16%, root mean square error (RMSE) = 0.27, and mean absolute percentage error (MAPE) = 5.71 for SBP prediction in normotensive subjects. Furthermore, R2 = 96.46%, RMSE = 0.19, and MAPE = 1.76 for SBP prediction and R2 = 95.44%, RMSE = 0.21, and MAPE = 2.78 for DBP prediction in hypertensive subjects using the PCA-LSSVM model. This assessment presents the importance and advantages posed by hybrid computing models for the prediction of variables in biomedical research studies.

Prediction of BP reactivity to talking using hybrid soft computing approaches.

High blood pressure (BP) is associated with an increased risk of cardiovascular diseases. Therefore, optimal precision in measurement of BP is appropriate in clinical and research studies. In this work, anthropometric characteristics including age, height, weight, body mass index (BMI), and arm circumference (AC) were used as independent predictor variables for the prediction of BP reactivity to talking. Principal component analysis (PCA) was fused with artificial neural network (ANN), adaptive neurofuzzy inference system (ANFIS), and least square-support vector machine (LS-SVM) model to remove the multicollinearity effect among anthropometric predictor variables. The statistical tests in terms of coefficient of determination (R (2)), root mean square error (RMSE), and mean absolute percentage error (MAPE) revealed that PCA based LS-SVM (PCA-LS-SVM) model produced a more efficient prediction of BP reactivity as compared to other models. This assessment presents the importance and advantages posed by PCA fused prediction models for prediction of biological variables.