Dynamic Learning Rate in Deep CNN Model for Metastasis Detection and Classification of Histopathology Images.

Diagnosis of different breast cancer stages using histopathology whole slide images (WSI) is the gold standard in determining the grade of tissue metastasis. Computer-aided diagnosis (CAD) assists medical experts as a second opinion tool in early detection to prevent further proliferation. The field of pathology has advanced so rapidly that it is possible to obtain high-quality images from glass slides. Patches from the region of interest in histopathology images are extracted and trained using artificial neural network models. The trained model primarily analyzes and predicts the histology images for the benign or malignant class to which it belongs. Classification of medical images focuses on the training of models with layers of abstraction to distinguish between these two classes with less false-positive rates. The learning rate is the crucial hyperparameter used during the training of deep convolutional neural networks (DCNN) to improve model accuracy. This work emphasizes the relevance of the dynamic learning rate than the fixed learning rate during the training of networks. The dynamic learning rate varies with preset conditions between the lower and upper boundaries and repeats at different iterations. The performance of the model thus improves and attains comparatively high accuracy with fewer iterations.

Fiber optic sensor for the measurement of concentration of silica in water with dual wavelength probing.

The design, development, and characterization of a fiber optic evanescent wave based sensor with selectivity suitable for concentration measurement are presented. The sensor that is made up of a step index plastic multimode fiber can be used for the measurement of silica in water. Generally evanescent wave fiber optic sensors employ a single source and detector that show change in output optical power irrespective of the interacting species, i.e., they lack selectivity. This design employing two sources provides excellent selectivity, and the differential arrangement further enhances sensitivity and repeatability. Advantages of this design include the use of inexpensive and easily available light-emitting diode sources that match the analytical wavelength of the samples. The use of dual wavelength probing topology enhances the selectivity, sensitivity, and repeatability, which cannot be achieved by single source evanescent wave fiber optic sensors.