Breast cancer has the highest fatality for women compared with other types of cancer. Generally, early diagnosis of cancer is crucial to increase the chances of successful treatment. Early diagnosis is possible through physical examination, screening, and obtaining a biopsy of the dubious area. In essence, utilizing histopathology slides of biopsies is more efficient than using typical screening methods. Nevertheless, the diagnosing process is still tiresome and is prone to human error during slide preparation, such as when dyeing and imaging. Therefore, a novel method is proposed for diagnosing breast cancer into benign or malignant in a magnification‐specific binary (MSB) classification. Besides, the introduced method classifies each type into four subclasses in a magnification‐specific multi‐category (MSM) fashion. The proposed method involves normalizing the hematoxylin and eosin stains to enhance colour separation and contrast. Then, two types of novel features—deep and shallow features—are extracted using two deep structure networks based on DenseNet and Xception. Finally, a multi‐classifier method based on the maximum value is utilized to achieve the best performance. The proposed method is evaluated using the BreakHis histopathology data set, and the results in terms of diagnostic accuracy are promising, achieving 99% and 92% in terms of MSB and MSM, respectively, compared with recent state‐of‐the‐art methods reported in the survey conducted by Benhammou on the BreakHis data set using deep learning and texture‐based models.

The authors propose a new method for sclera quality measure and segmentation under relaxed imaging constraints. In particular, for sclera image, they propose a new quality measure approach based on a focus measure. In addition, they propose a new fusion method for sclera segmentation which uses pixel properties of both the sclera area and the skin around the eye. Furthermore, sclera template rotation alignment and distance scaling methods are proposed to minimise the error rates when noisy eye images are captured at‐a‐distance and on‐the‐move, together with overcoming head pose rotation. Then, a performance analysis on exploited eye images using the Excellent, the Good, the Bad, and the Ugly (EGBU) classification technique for image quality is used to evaluate system performance. Eye images captured under relaxed imaging constraints using four camera devices within the UBIRIS.v2 and MICHE mobile databases are utilised to evaluate the proposed sclera recognition system, with the UBIRIS.v1 database as a reference. Results in terms of sclera image quality measure and sclera segmentation are promising and describe the effect and challenges of using relaxed imaging conditions on sclera recognition system.

An accurate tumour segmentation in brain images is a complicated task due to the complext structure and irregular shape of the tumour. In this letter, our contribution is twofold: (1) a lightweight brain tumour segmentation network (LBTS‐Net) is proposed for a fast yet accurate brain tumour segmentation; (2) transfer learning is integrated within the LBTS‐Net to fine‐tune the network and achieve a robust tumour segmentation. To the best of knowledge, this work is amongst the first in the literature which proposes a lightweight and tailored convolution neural network for brain tumour segmentation. The proposed model is based on the VGG architecture in which the number of convolution filters is cut to half in the first layer and the depth‐wise convolution is employed to lighten the VGG‐16 and VGG‐19 networks. Also, the original pixel‐labels in the LBTS‐Net are replaced by the new tumour labels in order to form the classification layer. Experimental results on the BRATS2015 database and comparisons with the state‐of‐the‐art methods confirmed the robustness of the proposed method achieving a global accuracy and a Dice score of 98.11% and 91%, respectively, while being much more computationally efficient due to containing almost half the number of parameters as in the standard VGG network.

Pneumonia is a severe infection that affects the lungs due to viral or bacterial infections such as the novel COVID-19 virus resulting in mild to critical health conditions. One way to diagnose pneumonia is to screen prospective patient’s lungs using either a Computed Tomography (CT) scan or chest X-ray. To help radiologists in processing a large amount of data especially during pandemics, and to overcome some limitations in deep learning approaches, this paper introduces a new approach that utilizes a few light-weighted densely connected bottleneck residual block features to extract rich spatial information. Then, shrinking data batches into a single vector using four efficient methods. Next, an adaptive weight setup is proposed utilizing Adaboost ensemble learning which adaptively sets weight for each classifier depending on the scores generated to achieve the highest true positive rates while maintaining low negative rates. The proposed method is evaluated using the Kaggle chest X-ray public dataset and attained an accuracy of 99.6% showing superiority to other deep networks-based pneumonia diagnosis methods.

Background: Preprocessing pipelines developed for retinal fundus photography are routinely transferred to other vascular imaging modalities without validation. Methods: Seven preprocessing stages from the retinal literature were assembled into a rule-based chain gated by image-quality descriptors calibrated from training-set percentiles. A ten-condition ablation (the complete chain, seven chains each omitting one stage, and two minimal chains of two stages each) with five-fold cross-validation was run on VessMAP, 100 confocal images of mouse cortical vasculature, and six conditions were repeated on the DRIVE and CHASE_DB1 fundus datasets. Results: On VessMAP, only removing intensity normalisation significantly reduced Dice ($-0.017$, paired Wilcoxon $p<{10}^{-4}$); removing local contrast normalisation raised it by 0.019, and two-stage chains exceeded the complete chain by 0.025–0.027. Removing local contrast normalisation or denoising, or reducing the chain to two stages, also improved Dice on both fundus datasets. With thresholds fitted on training folds only, VessMAP Dice was $0.882\pm 0.006$, within the inter-annotator interval [0.864, 0.891], and a 59,034-parameter network came within 0.012 Dice of a 20.9-million-parameter model. Conclusions: The complete chain was not the best configuration on any dataset, including the fundus datasets its stages came from; preprocessing should be validated stage by stage on the target data.