Acute decompensation (AD) of cirrhosis may present without acute-on-chronic liver failure (ACLF) (AD-No ACLF), or with ACLF (AD-ACLF), defined by organ failure(s). Herein, we aimed to analyze and characterize the precipitants leading to both of these AD phenotypes. The multicenter, prospective, observational PREDICT study (NCT03056612) included 1,273 non-electively hospitalized patients with AD (No ACLF = 1,071; ACLF = 202). Medical history, clinical data and laboratory data were collected at enrolment and during 90-day follow-up, with particular attention given to the following characteristics of precipitants: induction of organ dysfunction or failure, systemic inflammation, chronology, intensity, and relationship to outcome. Among various clinical events, 4 distinct events were precipitants consistently related to AD: proven bacterial infections, severe alcoholic hepatitis, gastrointestinal bleeding with shock and toxic encephalopathy. Among patients with precipitants in the AD-No ACLF cohort and the AD-ACLF cohort (38% and 71%, respectively), almost all (96% and 97%, respectively) showed proven bacterial infection and severe alcoholic hepatitis, either alone or in combination with other events. Survival was similar in patients with proven bacterial infections or severe alcoholic hepatitis in both AD phenotypes. The number of precipitants was associated with significantly increased 90-day mortality and was paralleled by increasing levels of surrogates for systemic inflammation. Importantly, adequate first-line antibiotic treatment of proven bacterial infections was associated with a lower ACLF development rate and lower 90-day mortality. This study identified precipitants that are significantly associated with a distinct clinical course and prognosis in patients with AD. Specific preventive and therapeutic strategies targeting these events may improve outcomes in patients with decompensated cirrhosis. Acute decompensation (AD) of cirrhosis is characterized by a rapid deterioration in patient health. Herein, we aimed to analyze the precipitating events that cause AD in patients with cirrhosis. Proven bacterial infections and severe alcoholic hepatitis, either alone or in combination, accounted for almost all (96-97%) cases of AD and acute-on-chronic liver failure. Whilst the type of precipitant was not associated with mortality, the number of precipitant(s) was. This study identified precipitants that are significantly associated with a distinct clinical course and prognosis of patients with AD. Specific preventive and therapeutic strategies targeting these events may improve patient outcomes.

Glioma is the most common type of tumor in humans originating in the brain. According to the World Health Organization, gliomas can be graded on a four-stage scale, ranging from the most benign to the most malignant. The grading of these tumors from image information is a far from trivial task for radiologists and one in which they could be assisted by machine-learning-based decision support. However, the machine learning analytical pipeline is also fraught with perils stemming from different sources, such as inadvertent data leakage, adequacy of 2D image sampling, or classifier assessment biases. In this paper, we analyze a glioma database sourced from multiple datasets using a simple classifier, aiming to obtain a reliable tumor grading and, on the way, we provide a few guidelines to ensure such reliability. Our results reveal that by focusing on the tumor region of interest and using data augmentation techniques we significantly enhanced the accuracy and confidence in tumor classifications. Evaluation on an independent test set resulted in an AUC-ROC of 0.932 in the discrimination of low-grade gliomas from high-grade gliomas, and an AUC-ROC of 0.893 in the classification of grades 2, 3, and 4. The study also highlights the importance of providing, beyond generic classification performance, measures of how reliable and trustworthy the model’s output is, thus assessing the model’s certainty and robustness.

This chapter delves into the pivotal role of transparency within artificial intelligence (AI)-based applications, emphasizing its importance for reliability, accountability, and ensuring the ethical usage of AI targeting healthcare contexts. The chapter examines four dimensions of transparency: data, algorithmic, decision-making, and AI solution, and identifies problems and limitations in achieving them based on real-world digital health use cases. Current efforts and recommended strategies aiming at boosting transparency are discussed, emphasizing the need to define the objectives of transparency, the circumstances under which it should be applied, and the rationale behind it. This chapter advocates for collaborative efforts from stakeholders throughout the healthcare ecosystem to prioritize and implement transparent AI systems for the benefit of patients and society.

Machine Learning is entering a phase of maturity, but its medical applications still lag behind in terms of practical use. The field of oncological radiology (and neuro-oncology in particular) is at the forefront of these developments, now boosted by the success of Deep-Learning methods for the analysis of medical images. This paper reviews in detail some of the most recent advances in the use of Deep Learning in this field, from the broader topic of the development of Machine-Learning-based analytical pipelines to specific instantiations of the use of Deep Learning in neuro-oncology; the latter including its use in the groundbreaking field of ultra-low field magnetic resonance imaging.

Background/Objectives: Adult-type diffuse gliomas (ADGs) are aggressive brain tumors with highly variable clinical outcomes. Prognostic assessment directly from routine hematoxylin and eosin (H&E)-stained whole-slide images (WSIs) could support personalized decision-making. Methods: To our knowledge, this is the first systematic benchmarking of pathology foundation models (FMs) and multiple instance learning (MIL) methods for overall survival (OS) risk stratification in ADGs under the 2021 World Health Organization (WHO) classification. A retrospective, multi-institutional cohort of 817 subjects from The Cancer Genome Atlas (TCGA) was analyzed. Following tissue segmentation and patch-level curation, we evaluated 77 pathology FM-MIL combinations (11 FMs × 7 MIL methods), together with seven ResNet-50 baseline configurations (84 encoder-MIL configurations in total), using patient-level 10-fold cross-validation. A perturbation-based evaluation assessed attention mechanism faithfulness. Results: UNI2 paired with MambaMIL achieved the highest cross-validated concordance index (C-index = 0.77), whereas its time-dependent AUC was modest (0.60–0.61 at 12–36 months), and several combinations performed comparably. Kaplan–Meier analysis demonstrated significant separation between high- and low-risk groups (log-rank p = 9.16 × 10⁻¹⁷; hazard ratio [HR] = 2.90, 95% confidence interval [CI]: 2.23–3.76), with a median OS of 16.6 months (95% CI: 15.0–20.4) versus 63.5 months (95% CI: 43.9-NR), respectively. Tertile stratification confirmed a monotonic OS gradient from 94.5 months to 14.7 months (p = 3.25 × 10⁻¹⁷). Risk stratification within individual WHO 2021 types was not significant. Most domain-specific FMs outperformed the ImageNet-pretrained baseline, and the removal of highly attended patches reduced performance, supporting attention faithfulness. Conclusions: The AI-derived risk score remained associated with OS after adjustment for individual clinical and molecular variables, but independent prognostic value was not demonstrated in the fully adjusted model (HR = 1.32, 95% CI: 0.94–1.85, p = 0.104), and its discrimination largely reflects differences between WHO 2021 types. This benchmarking framework provides guidance for FM-MIL design in computational pathology applications targeting survival prediction; external validation is required before clinical application.