Abstract
Lung cancer is still the leading cause of cancer death in the world. For this reason, novel approaches for early and more accurate diagnosis are needed. Computer-aided decision (CAD) can be an interesting option for a noninvasive tumour characterisation based on thoracic computed tomography (CT) image analysis. Until now, radiomics have been focused on tumour features analysis, and have not considered the information on other lung structures that can have relevant features for tumour genotype classification, especially for epidermal growth factor receptor (EGFR), which is the mutation with the most successful targeted therapies. With this perspective paper, we aim to explore a comprehensive analysis of the need to combine the information from tumours with other lung structures for the next generation of CADs, which could create a high impact on targeted therapies and personalised medicine. The forthcoming artificial intelligence (AI)-based approaches for lung cancer assessment should be able to make a holistic analysis, capturing information from pathological processes involved in cancer development. The powerful and interpretable AI models allow us to identify novel biomarkers of cancer development, contributing to new insights about the pathological processes, and making a more accurate diagnosis to help in the treatment plan selection.

Abstract
Total white blood cells (WBC) count is an important indication for infection diagnosis, in both human and veterinary medicine. State-of-the-art WBC counts are performed by flow cytometry combined with light scattering or impedance measurements, in the clinical analysis laboratory. These technologies are complex and difficult to be miniaturized into a portable point-of-care (POC) system. Spectroscopy is one of the most powerful technologies for POC miniaturization due to its capacity to analyze low sample quantities, little to no sample preparation, and ‘real-time’ results. WBC is in the proportion of 1:1000 to red blood cells (RBC), and the latter dominate visible-near infrared (Vis-NIR) information due to their large quantities and hemoglobin absorbance. WBC are difficult to be detected by traditional spectral analysis because their information is contained within the interference of hemoglobin bands. Herein, we perform a feasibility study for the direct detection of WBC counts in canine blood by Vis-NIR spectroscopy for veterinary applications, benchmarking current chemometrics techniques with self-learning artificial intelligence—a new advanced method for high-accuracy quantification from spectral information. Results show that total WBC counts can be detected by Vis-NIR spectroscopy to an average detection limit of 7.8 ×109 cells/L, with an R2 of 0.9880 between impedance flow cytometry analysis and spectral quantification. This result opens new possibilities for reagent-less POC technology in infection diagnosis. As WBC counts in dogs range from 5 to 45 ×109 cells/L, the detection limit obtained in this research allows concluding that the combined use of spectroscopy with this SL-AI new algorithm is a step towards the existence of portable and miniaturized Spectral POC hemogram analysis.

Abstract
Due to the huge mortality rate of lung cancer, there is a strong need for developing solutions that help with the early diagnosis and the definition of the most appropriate treatment. In the particular case of target therapy, effective genotyping of the tumor is fundamental since this treatment uses targeted drugs that can induce death in cancer cells. The biopsy is the traditional method to assess the genotype information but it is extremely invasive and painful. Medical imaging is a valuable alternative to biopsies, considering the potential to extract imaging features correlated with specific genomic alterations. Regarding the limitations of single model approaches for gene mutation status predictions, ensemble strategies might bring valuable benefits by combining the strengths and weaknesses of the aggregated methods. This preliminary work aims to provide further advances in the radiogenomics field by studying the use of ensemble methods to predict the Epidermal Growth Factor Receptor (EGFR) mutation status in lung cancer. The best result obtained for the proposed ensemble approach was an AUC of 0.706 (± 0.122). However, the ensemble did not outperform the single models with AUC values of 0.712 (± 0.119) for Logistic Regression, 0.711 (± 0.119) for Support Vector Machine and 0.712 (± 0.120) for Elastic Net. The high correlation found on the decisions of each single model might be a plausible explanation for this behavior, which caused the ensemble to misclassify the same examples as the single models.

Abstract
Chest radiography is one of the most ubiquitous medical imaging exams used for the diagnosis and follow-up of a wide array of pathologies. However, chest radiography analysis is time consuming and often challenging, even for experts. This has led to the development of numerous automatic solutions for multipathology detection in chest radiography, particularly after the advent of deep learning. However, the black-box nature of deep learning solutions together with the inherent class imbalance of medical imaging problems often leads to weak generalization capabilities, with models learning features based on spurious correlations such as the aspect and position of laterality, patient position, equipment and hospital markers. In this study, an automatic method based on a YOLOv3 framework was thus developed for the detection of markers and written labels in chest radiography images. It is shown that this model successfully detects a large proportion of markers in chest radiography, even in datasets different from the training source, with a low rate of false positives per image. As such, this method could be used for performing automatic obscuration of markers in large datasets, so that more generic and meaningful features can be learned, thus improving classification performance and robustness.

Abstract
SARS-CoV-2 has presented direct and indirect challenges to the scientific community. One of the most prominent indirect challenges advents from the mandatory use of face masks in a large number of countries. Face recognition methods struggle to perform identity verification with similar accuracy on masked and unmasked individuals. It has been shown that the performance of these methods drops considerably in the presence of face masks, especially if the reference image is unmasked. We propose FocusFace, a multi-task architecture that uses contrastive learning to be able to accurately perform masked face recognition. The proposed architecture is designed to be trained from scratch or to work on top of state-of-the-art face recognition methods without sacrificing the capabilities of a existing models in conventional face recognition tasks. We also explore different approaches to design the contrastive learning module. Results are presented in terms of masked-masked (MM) and unmasked-masked (U-M) face verification performance. For both settings, the results are on par with published methods, but for M-M specifically, the proposed method was able to outperform all the solutions that it was compared to. We further show that when using our method on top of already existing methods the training computational costs decrease significantly while retaining similar performances. The implementation and the trained models are available at GitHub.

Abstract
Epilepsy is a major neurological disorder affecting approximately 1% of the world population, where seizure semiology is an essential tool for clinical evaluation of seizures. This includes qualitative visual inspection of videos from the seizures in epilepsy monitoring units by epileptologists. In order to support this clinical diagnosis process, promising deep learning-based systems were proposed. However, these indicate that video datasets of epileptic seizures are still rare and limited in size. In order to enable the full potential of AI systems for epileptic seizure diagnosis support and research, a novel collaborative development framework is proposed for a scalable DL-assisted clinical research and diagnosis support of epileptic seizures. The designed cloud-based approach integrates our deployed and tested NeuroKinect data acquisition pipeline into an MLOps framework to scale data set extension and analysis to a multi-clinical utilization. The proposed development framework incorporates an MLOps approach, to ensure convenient collaboration between clinicians and data scientists, providing continuous advantages to both user groups. It addresses methods for efficient utilization of HW, SW and human resources. In the future, the system is going to be expanded with several AI-based tools. Such as DL-based automated 3D motion capture (MoCap), 3D movement analysis support, quantitative seizure semiology analysis tools, video-based MOI and seizure classification. © 2021 IEEE