Hierarchical skin lesion image classification with prototypical decision tree.

Traditional disease classification models often disregard the clinical significance of misclassifications and lack interpretability. To overcome these challenges, we propose a hierarchical prototypical decision tree (HPDT) for skin lesion classification. HPDT combines prototypical networks and decision trees, leveraging a class hierarchy to guide interpretable predictions from general to specific categories.

Aortic pressures within a giant right coronary artery aneurysm.

Background: Giant coronary artery aneurysms are rare conditions with potentially devastating consequences. We report a case of the largest documented giant right coronary artery (RCA) aneurysm to date.

Case Presentation: A 57-year-old male patient visited our outpatient clinic for abdominal pain and exertional dyspnea.

Metal (Au, Pt, Pd, Ni) Bis(dithiolene) complexes as dual-action agents combating cancer and trypanosomatid infections.

Cancer and infection diseases pose severe threats to public health worldwide stressing the need for more effective and efficient treatments. Thus, the search for broad-spectrum activity drugs seems justifiable and urgent. Herein, we investigate the anticancer and antitrypanosomatid (anti-Trypanosoma cruzi) activities of eight monoanionic metal bis(dithiolene) complexes, [PhP][M(R-thiazdt)] with M = Au, Pt, Pd, Ni, containing N-alkyl-1,3-thiazoline-2-thione dithiolene ligands (R-thiazdt) with different alkyl groups (R = Et, tBu).

Mixed-ligand, radical, gold bis(dithiolene) complexes: from single-component conductors to controllable NIR-II absorbers.

Neutral radical bis(dithiolene) gold complexes [Au(dt)]˙ are known to exhibit a strong absorption in the 1400-2000 nm NIR absorption range. Here, we demonstrate that the NIR signature of mixed-ligand bis(dithiolene) gold complexes [Au(dt)(dt)]˙ associating two different dithiolene, dt and dt, is found at higher energy, out of the range of the homoleptic analogs [Au(dt)]˙ and [Au(dt)]˙, in the looked-after NIR-II 1000-1400 nm absorption range. An efficient synthetic approach towards precursor mixed-ligand monoanionic gold bis(dithiolene) complexes [Au(dt)(dt)] is reported.

AI models collapse when trained on recursively generated data.

Stable diffusion revolutionized image creation from descriptive text. GPT-2 (ref. ), GPT-3(.

Structure-activity relationship of anticancer and antiplasmodial gold bis(dithiolene) complexes.

Monoanionic gold bis(dithiolene) complexes were recently shown to display activity against ovarian cancer cells, Gram-positive bacteria, strains and the rodent malaria parasite, . To date, only monoanionic gold(III) bis(dithiolene) complexes with a thiazoline backbone substituted with small alkyl chains have been evaluated for biomedical applications. We now analyzed the influence of the length and the hydrophobicity hydrophilicity of these complexes' alkyl chain on their anticancer and antiplasmodial properties.

Detecting hallucinations in large language models using semantic entropy.

Large language model (LLM) systems, such as ChatGPT or Gemini, can show impressive reasoning and question-answering capabilities but often 'hallucinate' false outputs and unsubstantiated answers. Answering unreliably or without the necessary information prevents adoption in diverse fields, with problems including fabrication of legal precedents or untrue facts in news articles and even posing a risk to human life in medical domains such as radiology. Encouraging truthfulness through supervision or reinforcement has been only partially successful.

Radical and diradical states of bis(molybdenocene dithiolene) complexes.

The synthesis and characterization of two bis(dithiolene) proligands involving heteroatomic linkers such as 1,4-dithiine and dihydro-1,4-disiline between the two protected dithiolene moieties are described. Two bimetallic complexes involving these heteroatomic bridges between two redox active bis(cyclopentadienyl)molybdenum dithiolene moieties have been synthesized and characterized by electrochemistry, spectroelectrochemistry, and their properties rationalized with (TD-)DFT. Cyclic voltammetry experiments show sequential oxidation of the two redox centers with Δ values between successive one-electron transfers varying according to the nature of the bridge.

Long-Term Pulmonary Damage in Surviving Antitoxin-Treated Mice following a Lethal Ricin Intoxication.

Ricin, a highly potent plant-derived toxin, is considered a potential bioterrorism weapon due to its pronounced toxicity, high availability, and ease of preparation. Acute damage following pulmonary ricinosis is characterized by local cytokine storm, massive neutrophil infiltration, and edema formation, resulting in respiratory insufficiency and death. A designated equine polyclonal antibody-based (antitoxin) treatment was developed in our laboratory and proved efficacious in alleviating lung injury and increasing survival rates.

Machine learning for functional protein design.

Recent breakthroughs in AI coupled with the rapid accumulation of protein sequence and structure data have radically transformed computational protein design. New methods promise to escape the constraints of natural and laboratory evolution, accelerating the generation of proteins for applications in biotechnology and medicine. To make sense of the exploding diversity of machine learning approaches, we introduce a unifying framework that classifies models on the basis of their use of three core data modalities: sequences, structures and functional labels.

Single versus bilateral internal thoracic artery grafting in patients with impaired renal function.

Objective: The optimal strategy for surgical revascularization in patients with impaired renal function is inconclusive. We compared early and late outcomes between bilateral internal thoracic artery (BITA) and single ITA (SITA) grafting in patients with renal dysfunction.

Methods: This is a retrospective analysis of all the patients with multivessel disease and impaired renal function (estimated glomerular filtration rate <60mL/min/1.

Gaps in pediatric oncologic end of life care as recounted by the bereaved parents in Isarel.

Purpose: This study investigated parents' perception of their needs and those of their children with cancer at the end-of-life period, including unmet needs and their expectations regarding providers.

Design And Methods: This cross-sectional study involved 26 parents recruited from three pediatric hematology-oncology wards in Israel who completed demographic and medical questionnaires of the child, and a parental needs questionnaire based on The Needs Assessment of Family Caregivers-Cancer questionnaire, following the death of their child.

Findings: Parents expressed needs related to medical care, including pain management, decision-making, and finding optimal treatment options for their children.

Facile Fabrication of NIR-Responsive Alginate/CMC Hydrogels Derived through IEDDA Click Chemistry for Photothermal-Photodynamic Anti-Tumor Therapy.

Novel chemically cross-linked hydrogels derived from carboxymethyl cellulose (CMC) and alginate (Alg) were prepared through the utilization of the norbornene (Nb)-methyl tetrazine (mTz) click reaction. The hydrogels were designed to generate reactive oxygen species (ROS) from an NIR dye, indocyanine green (ICG), for combined photothermal and photodynamic therapy (PTT/PDT). The cross-linking reaction between Nb and mTz moieties occurred via an inverse electron-demand Diels-Alder chemistry under physiological conditions avoiding the need for a catalyst.

ProteinGym: Large-Scale Benchmarks for Protein Design and Fitness Prediction.

Predicting the effects of mutations in proteins is critical to many applications, from understanding genetic disease to designing novel proteins that can address our most pressing challenges in climate, agriculture and healthcare. Despite a surge in machine learning-based protein models to tackle these questions, an assessment of their respective benefits is challenging due to the use of distinct, often contrived, experimental datasets, and the variable performance of models across different protein families. Addressing these challenges requires scale.

ProteinNPT: Improving Protein Property Prediction and Design with Non-Parametric Transformers.

Protein design holds immense potential for optimizing naturally occurring proteins, with broad applications in drug discovery, material design, and sustainability. However, computational methods for protein engineering are confronted with significant challenges, such as an expansive design space, sparse functional regions, and a scarcity of available labels. These issues are further exacerbated in practice by the fact most real-life design scenarios necessitate the simultaneous optimization of multiple properties.

Learning from prepandemic data to forecast viral escape.

Effective pandemic preparedness relies on anticipating viral mutations that are able to evade host immune responses to facilitate vaccine and therapeutic design. However, current strategies for viral evolution prediction are not available early in a pandemic-experimental approaches require host polyclonal antibodies to test against, and existing computational methods draw heavily from current strain prevalence to make reliable predictions of variants of concern. To address this, we developed EVEscape, a generalizable modular framework that combines fitness predictions from a deep learning model of historical sequences with biophysical and structural information.

Mind the Gap-A Perspective on Strategies for Protecting against Bacterial Infections during the Period from Infection to Eradication.

The emergence of antibiotic-resistant bacteria is a pressing public health concern, highlighting the need for alternative approaches to control bacterial infections. Promising approaches include the development of therapeutic vaccines and the utilization of innate immune activation techniques, which may prove useful in conjunction with antibiotics, as well as other antibacterial modalities. However, innate activation should be fast and self- or actively- contained to prevent detrimental consequences.

Clinically Evaluated COVID-19 Drugs with Therapeutic Potential for Biological Warfare Agents.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) outbreak resulted in hundreds of millions of coronavirus cases, as well as millions of deaths worldwide. Coronavirus Disease 2019 (COVID-19), the disease resulting from exposure to this pathogen, is characterized, among other features, by a pulmonary pathology, which can progress to "cytokine storm", acute respiratory distress syndrome (ARDS), respiratory failure and death. Vaccines are the unsurpassed strategy for prevention and protection against the SARS-CoV-2 infection.

A Bimetallic Benzene-1,2,4,5-Tetrathiolate (btt) Molybdenocene Complex Cp Mo(btt)MoCp : Radical and Diradical States.

Benzene-1,2,4,5-tetrathiolate (btt) has been used as a bridging ligand to prepare a redox active (molybdenocene dithiolene)-based bimetallic complex Cp Mo(btt)MoCp , which exhibits four successive electron transfers up to the tetracation. Spectro-electrochemical investigations together with DFT and TD-DFT calculations evidence that the two electroactive MoS C metallacycles are electronically coupled in the monocationic as in the dicationic state. Two salts of the dication [Cp Mo(btt)MoCp ] have been structurally characterized with PF and HSO counterions, showing different chair or boat conformations associated with variable folding angles of the two MoS C metallacycles along the S-S hinge.

Heart Valve Surgery in Antiphospholipid Syndrome Patients-Morbidity and Mortality.

Objectives: To assess valve surgery outcomes in antiphospholipid syndrome (APS).

Methods: A retrospective study assessing complications and mortality rate and possible factors associated with adverse outcomes of APS patients undergoing valve surgery in two tertiary medical centers.

Results: Twenty-six APS patients (median age at surgery 47.

QU-BraTS: MICCAI BraTS 2020 Challenge on Quantifying Uncertainty in Brain Tumor Segmentation - Analysis of Ranking Scores and Benchmarking Results.

Deep learning (DL) models have provided state-of-the-art performance in various medical imaging benchmarking challenges, including the Brain Tumor Segmentation (BraTS) challenges. However, the task of focal pathology multi-compartment segmentation (e.g.

Sex differences in heart failure patients assessed by combined echocardiographic and cardiopulmonary exercise testing.

Background: We aimed to test the differences in peak VO between males and females in patients diagnosed with heart failure (HF), using combined stress echocardiography (SE) and cardiopulmonary exercise testing (CPET).

Methods: Patients who underwent CPET and SE for evaluation of dyspnea or exertional intolerance at our institution, between January 2013 and December 2017, were included and retrospectively assessed. Patients were divided into three groups: HF with preserved ejection fraction (HFEF), HF with mildly reduced or reduced ejection fraction (HFEF/HFEF), and patients without HF (control).

Successful conservative management of left ventricular assist device candidates.

Aims: Clinical trials comparing LVADs vs. conservative therapy were performed before the availability of novel medications or used suboptimal medical therapy. This study aimed to report that long-term stabilization of patients entering a left ventricular assist device (LVAD) programme is possible with the use of aggressive conservative therapy.