The accurate classification of skin lesions, particularly melanoma, is vital for the early detection and effective treatment of skin cancer. Although deep learning models such as convolutional neural networks (CNNs) have achieved remarkable success in dermoscopic image analysis, they often overlook valuable structured metadata (e.g., patient demographics, lesion location, and type) that provide essential diagnostic context. We present a graph-driven multimodal framework that jointly models visual and metadata information for skin lesion classification. Our approach uses a frozen CNN backbone to extract deep visual representations via dual pooling (mean and max), which are concatenated with encoded metadata and partitioned into subspaces. These subspaces are treated as nodes within a graph, where a graph neural network (GNN) captures intra-sample dependencies between feature subspaces and clinical attributes to refine lesion representations. Experiments on four public benchmarks: ISIC2024, HAM10000, PAD-UFES-20, and HIBA, demonstrating consistent performance gains over several state-of-the-art (SOTA) approaches, with relative accuracy improvements ranging from +0.5% to +8.8% in datasets. The results highlight the potential of graph-based modeling of metadata and image features to build more robust and clinically informed skin cancer classifiers.

Machine unlearning is the process of efficiently removing specific information from a trained machine learning model without retraining from scratch. Existing unlearning methods, which often provide provable guarantees, typically involve retraining a subset of model parameters based on a forget set. While these approaches show promise in certain scenarios, their underlying assumptions are often challenged in real-world applications -- particularly when applied to generative models. Furthermore, updating parameters using these unlearning procedures often degrades the general-purpose capabilities the model acquired during pre-training. Motivated by these shortcomings, this paper considers the paradigm of inference time unlearning -- wherein, the generative model is equipped with an (approximately correct) verifier that judges whether the model's response satisfies appropriate unlearning guarantees. This paper introduces a framework that iteratively refines the quality of the generated responses using feedback from the verifier without updating the model parameters. The proposed framework leverages conformal prediction to reduce computational overhead and provide distribution-free unlearning guarantees. This paper's approach significantly outperforms existing state-of-the-art methods, reducing unlearning error by up to 93% across challenging unlearning benchmarks.

Concept erasure is the task of erasing information about a concept (e.g., gender or race) from a representation set while retaining the maximum possible utility -- information from original representations. Concept erasure is useful in several applications, such as removing sensitive concepts to achieve fairness and interpreting the impact of specific concepts on a model's performance. Previous concept erasure techniques have prioritized robustly erasing concepts over retaining the utility of the resultant representations. However, there seems to be an inherent tradeoff between erasure and retaining utility, making it unclear how to achieve perfect concept erasure while maintaining high utility. In this paper, we offer a fresh perspective toward solving this problem by quantifying the fundamental limits of concept erasure through an information-theoretic lens. Using these results, we investigate constraints on the data distribution and the erasure functions required to achieve the limits of perfect concept erasure. Empirically, we show that the derived erasure functions achieve the optimal theoretical bounds. Additionally, we show that our approach outperforms existing methods on a range of synthetic and real-world datasets using GPT-4 representations.

Glucagon-like peptide-1 (GLP-1) is a crucial incretin hormone that regulates glucose homeostasis by enhancing insulin secretion, suppressing glucagon release, and delaying gastric emptying. While synthetic GLP-1 receptor agonists such as semaglutide have demonstrated efficacy in managing type 2 diabetes mellitus and obesity, their high cost, limited accessibility, and adverse effects have limited their applicability, necessitating the search for alternative therapeutic strategies. Peganum harmala (harmal), a traditional medicinal plant, has gained attention for its bioactive alkaloids, harmine, and harmaline, which have been shown to modulate key molecular pathways involved in GLP-1 secretion and insulin sensitization. These alkaloids enhance Akt phosphorylation (pS473-Akt), facilitating glucose transporter type 4 translocation and glucose uptake, while concurrently activating the nuclear factor erythroid 2-related factor 2 pathway, leading to increased antioxidant defenses and reduced oxidative stress in pancreatic β-cells and enteroendocrine L-cells. Furthermore, P. harmala alleviates insulin resistance by suppressing IRS-1 serine phosphorylation (pS307-IRS-1) and improving phosphoinositide 3-kinase/Akt signaling, thereby optimizing insulin receptor sensitivity and metabolic homeostasis. Despite these promising pharmacological properties, the poor solubility and rapid metabolism of harmine and harmaline pose challenges to their clinical application. Nanotechnology-based drug delivery systems, including liposomal encapsulation and polymeric nanoparticles, offer a potential solution to enhance bioavailability, prolong systemic circulation, and enable targeted delivery to GLP-1-secreting cells. This paper delves into the molecular mechanisms by which P. harmala stimulates GLP-1 secretion and improves insulin sensitivity, compares its effects with semaglutide, and highlights the potential role of nanotechnology in optimizing its therapeutic applications. By integrating traditional medicine with modern pharmaceutical advancements, P. harmala represents a promising, cost-effective, and sustainable approach to metabolic disorder management, warranting further investigation through pre-clinical and clinical studies.

Background MicroRNAs (miRNAs) are implicated in regulating obesity, but clinical trials have yielded conflicting results. Objective This study aimed to investigate the relationships between circulating levels of miRNA-221 and miRNA-222 in obesity and hypertension-related obesity in humans. This may shed light on the pathogenic pathways controlling obesity and provide noninvasive molecular indicators to identify and predict the disease. The relationships between circulating levels of the above miRNAs and variables related to adiposity and lipid profiles were further investigated. Patients and methods Using a quantitative real-time P technique, the expression levels of circulating miRNAs were determined in serum samples from 65 obese patients (35 without and 30 with hypertension) and 45 age-matched and sex-matched normal-weight individuals. Results and conclusion MiRNAs (221 and 222) have been shown to be differentially expressed in the sera of obese patients compared to controls. In addition, obese hypertensive patients were shown to have higher serum levels of miRNA-222 and miRNA-221 than healthy controls. Serum miRNA-221 was correlated with BMI, cholesterol, triglycerides, and low-density lipoprotein cholesterol. Thus, circulating levels of miRNA-221 and miRNA-222 inease in obesity and obesity-associated hypertension. These miRNAs may serve as markers for obesity and obesity-induced hypertension.

The development of advanced image-generative modalities has significantly improved digitalized histopathological diagnostics. Despite its limitations, hematoxylin and eosin (H&E) staining remains the gold standard for cancer diagnoses. However, the contrast in H&E-stained tissue specimens can be challenging to distinguish, necessitating more specific staining approaches. Immunohistochemistry (IHC) addresses this issue by employing antibodies that bind specifically to antigens in biological tissues. However, IHC is time-consuming, expensive, and labor-intensive. Anovel deep-learning-based approach is proposed, using a conditional diffusion-based model to generate virtually IHC-stained images from H&E images. The state-of-the-art methods address this image-to-image translation task by formulating it as a problem in generative adversarial networks (GANs), however, our proposed method demonstrates improved performance due to its stable training process. The results on a benchmark dataset show that our proposed method can overcome the limitations of the state-of-the-art staining methods such as CycleGAN and pix2pix with improved PSNR, SSIM and FID scores and closer visual quality to the ground truth IHC images.

When training transformers on graph-structured data, incorporating information about the underlying topology is crucial for good performance. Topological masking, a type of relative position encoding, achieves this by upweighting or downweighting attention depending on the relationship between the query and keys in the graph. In this paper, we propose to parameterise topological masks as a learnable function of a weighted adjacency matrix -- a novel, flexible approach which incorporates a strong structural inductive bias. By approximating this mask with graph random features (for which we prove the first known concentration bounds), we show how this can be made fully compatible with linear attention, preserving $\mathcal{O}(N)$ time and space complexity with respect to the number of input tokens. The fastest previous alternative was $\mathcal{O}(N \log N)$ and only suitable for specific graphs. Our efficient masking algorithms provide strong performance gains for image and point cloud data, including with $>30$k nodes.

Background: Obesity is a concerning health problem globally. Previous studies demonstrated the role of miR-15a and miRNA Let-7 in adipocyte development and in regulation of glucose metabolism.Blood samples of 65 obese patients (17 obese hypertensive patients, 24 obese diabetic patients, and 24 obese without complications) and 45 healthy volunteers were used as control.Fasting blood glucose, fasting insulin, lipid profile and HOMA-IR as well as miRNA-15a and miRNA Let-7 expression levels were assessed. Results: All tested biochemical parameters were significantly increased in all obese groups compared with the controls. The expression of miRNA 15a was significantly decreased in all obese groups in comparison with the controls especially the obese diabetic group, while the expression of miRNA Let-7 showed no significant difference between the studied groups with only a significant increase in obese hypertensive group. Conclusion: MiR15a level was reduced in the obese diabetic group, this finding highlights the role of miR15a in diabetes development. MiRNA Let-7 expression was decreased in diabetic obese patients while increased in hypertensive obese, suggesting its value in progression of diabetes and hypertension as obesity complications.

Actinomycetes found in symbiosis with marine sponges have the potential to yield new effective pharmaceuticals, such as antifungal, antibiotic, anticancer, and antiviral compounds. This study aims to identify bioactive metabolites from these actinomycetes. Biologically guided screening for antifungal and cytotoxic activity of isolated actinomycetes revealed that 37 species exhibited a broad spectrum of potency as antifungal, anticancer, antioxidants, and antibiotics. The promising isolate was phenotypically and genotypically identified as Streptomyces acrimycini strain MBS-HRS-EG (MYI1) (ID: PP477812.1). In vitro, assessment of strain MBS-HRS-EG exhibited a broad spectrum of antifungal potency against fungal pathogens such as Candida albicans, Candida tropicalis, Aspergillus Niger, Fusarium oxisporium, Fusarium solani, and Rhizoctonia solani by 12, 19, 16, 17, 21, and 18 mm, respectively. The fermentation broth of the selected strain has a broad-spectrum potency as an antibacterial effect against gram-positive pathogenic bacteria like Bacillus cereus, Staphylococcus aureus, and Streptococcus pyogenes by 16, 19, and 26 mm, respectively. Additionally, it showed activity against gram-negative bacteria such as Escherichia coli, Klebsiella Pneumonia, and Pseudomonas aeruginosa by 15, 19, and 20 mm, respectively. The results of the DPPH radical scavenging activities showed that the radical scavenging activity was 95.61%, closest to the reference ascorbic acid at 99.84%. Chemical profiling of the strain fermentation broth was conducted using UHPLCQTOF-MS/MS. The produced secondary metabolites were identified using MS DIAL. Confirmation of the UHPLC/QTOF MS analysis data for the bioactive actinomycetes metabolites was done by comparing their analysis data with identified compounds presented in the DNP database and available database of ordinary products to find the nearest matches for the detected metabolites. The analysis identified 21 bioactive metabolites, including macrolide antibiotics such as Natamycin, Strevertene A, Albocyclin, Bafilomycin D with antifungal activity, cyclic peptides like Maculosin, macrocyclic peptides such as Microsclerodermin D, angucyclic antibiotics like Landomycin H with antifungal, anticancer, and antibiotic properties, polyketides such as Delactonmycin with antiviral activity, and lipodepsipeptide like Viridamide A with antiprotozoal activity. The results concluded that the potent strain can be applied to produce various pharmaceutical products.

This study delves into the synergistic interplay among albumin, insulin, Klotho protein, and relevant medications in the pursuit of a concerted approach to halt aging. Emphasizing albumin's pivotal role in various physiological functions, including transportation and drug distribution, the research underscores its decline's correlation with aging-related cognitive implications. The intricate relationship between insulin and albumin, modulated by Foxo1, underscores its crucial significance. Groundbreaking experiments, utilizing unmodified serum albumin, demonstrate a remarkable increase in lifespan and enhanced physical capabilities, highlighting the potential of an integrative approach. The investigation extends to insulin sensitizers, Klotho protein, metformin, and SGLT2 inhibitors, collectively revealing promising anti-aging effects. The association between Klotho protein and albumin suggests a collaborative mechanism with implications for efficient transport and distribution. This research offers insights into a comprehensive, synergistic strategy harnessing the potential of these elements to counteract aging processes.

Triple-positive breast cancer, characterized by the concurrent overexpression of estrogen receptors (ER+), progesterone receptors (PR+), and human epidermal growth factor receptor 2 (HER2+), presents a significant clinical challenge in oncology. This particular subtype, distinguished by its aggressive behavior and propensity for metastasis, necessitates a comprehensive therapeutic approach. Current treatment modalities, primarily centered around targeted therapies, encounter obstacles, underscoring the imperative to explore alternative interventions. The emergence of Glucosodiene, grounded in Maher Akl's hypothesis regarding glucose mutation, introduces a promising avenue for therapeutic intervention. This innovative pharmacological agent exhibits efficacy in targeting the Warburg effect, a characteristic feature of tumors reliant on anaerobic glucose metabolism. A positron emission tomography (PET) scan conducted on a 36-year-old female patient following oral administration of Glucosodiene at a daily dosage of 100 ml over 15 consecutive days revealed encouraging findings, including regression of lesions in the left breast and a favorable response in axillary lymph nodes. Additionally, improvement was evident in the abdomino-pelvic region and musculoskeletal system, indicative of a partial metabolic response compared to prior imaging studies. Noteworthy reductions were observed in the number, size, and metabolic activity of osseous lesions, indicative of favorable disease progression. The mechanistic underpinnings of Glucosodiene position it as a versatile and impactful therapeutic option in the landscape of cancer management, offering promise for enhanced patient outcomes. THE TRIAL IS REGISTERED UNDER CLINICALTRIALS.GOV NUMBER NCT05957939.

Cancer is a complex genetic disease characterized by aberrant cellular behaviors, including uncontrolled growth, invasion, and metastasis. The development of personalized treatment strategies based on genomic profiling has led to improved outcomes. Recent scientific endeavors have focused on targeting cancer through metabolic approaches, capitalizing on the altered metabolic pathways in cancer cells. Glucosodiene polymer, a newly derived compound from glucose, has shown promising results in inhibiting glucose metabolism and modifying the tumor's microenvironment acidity. The Maher Akl Theory "Glucose Mutation" proposes a strategic approach to target cancerous tumors by inhibiting glucose metabolism and altering the tumor's microenvironment acidity using glucose isomer polymers. The goal is to disrupt the metabolic activity of the tumor and potentially modify and control the disease. This manuscript provides an overview of the metabolic vulnerabilities of cancer cells, evaluates the synthesis and chemical structure of glucosodiene, documents its safety, and explores its potential as a targeted therapy for cancer treatment. Additionally, a subset of successful clinical trials is presented, focusing on a case of successful treatment of triple-negative breast cancer (TNBC) with glucosodiene, and Medical Guidance and Integrated Therapeutic Approach: The Protocol of Glucose Mutation Theory via Glucosodiene and indication of Positive Tumor Lysis Syndrome "The potential mechanisms of action of glucosodiene in cancer, including its impact on glucose metabolism, modulation of signaling pathways, and immune-enhancing effects, are discussed.

This manuscript explores the potential of artificial intelligence (AI) to assist physicians in making informed fasting decisions for diabetic patients during Ramadan, a period that poses significant health challenges due to extended fasting hours. It focuses on two main concerns: the increased risk of hypoglycemia, particularly in those with type 1 diabetes or type 2 diabetes who are on insulin therapy or insulin secretagogues, and potential adverse drug interactions due to the complexity of managing diabetes and other comorbid conditions during fasting. Emphasizing the utility of AI for optimizing medication schedules, the manuscript highlights the significance of continuous glucose monitoring systems (CGMs) and references the PROFAST study to underline AI’s capability in assessing hypoglycemia risks. By showcasing AI’s role in enhancing diabetic care through risk prediction and decision-making support, the manuscript calls for expanded research and the ethical integration of AI technologies in healthcare, underscoring its capacity to provide personalized support and improve health outcomes during Ramadan.

Triple-positive breast cancer, characterized by the overexpression of estrogen receptors (ER+), progesterone receptors (PR+), and human epidermal growth factor receptor 2 (HER2+), poses a formidable challenge in oncology. This subtype, known for its aggressive nature and metastatic potential, requires a comprehensive therapeutic strategy. Current approaches, predominantly involving targeted therapies, face challenges, necessitating the exploration of alternative interventions. The emergence of Glucosodiene, rooted in Maher Akl's theory on glucose mutation, introduces a promising avenue for treatment. This innovative drug demonstrates efficacy in targeting the Warburg effect prevalent in tumors dependent on anaerobic glucose metabolism. A case study involving a 35-year-old woman with stage II triple-positive breast cancer showcases Glucosodiene's impact, revealing a complete absence of active lesions post-treatment. The results indicate its potential as a primary or secondary therapy, complementing traditional treatment protocols. Glucosodiene's mechanism of action positions it as a versatile and impactful option in the realm of cancer therapeutics, offering hope in the pursuit of improved patient outcomes. THE TRIAL IS REGISTERED UNDER CLINICALTRIALS.GOV NUMBER NCT05957939.

The knowledge and awareness of health professionals and patients surrounding black box warnings are still gaping. Our target is finding solutions to prevent potential side effects and severe interactions. Black box warnings are the strongest drug safety warnings issued by regulatory authorities, highlighting the potential risks associated with specific medications. However, there is often limited understanding and awareness of these warnings among both healthcare providers and patients, which can lead to inadequate risk management and patient safety. Therefore, closing knowledge gaps and proposing strategies for improving awareness and reducing adverse events related to medications with black box warnings is crucial. As we witness an unprecedented surge in drug discoveries, especially biological drugs for immunological disorders and cancer therapy, growing dramatically and receiving approvals by the FDA, the landscape of healthcare is rapidly evolving. With the remarkable advancements in immunotherapy, particularly the proliferation of monoclonal antibody drugs, we are witnessing a paradigm shift in the treatment of various conditions. However, this surge in drug approvals has brought forth a crucial concern - the increasing prevalence of Black Box Warnings associated with many of these drugs. A Black Box Warning, recognized as the most critical indication of potential serious side effects by the FDA, is typically issued post-marketing. This revelation highlights a significant gap in our understanding, particularly in terms of knowledge and adherence to these warnings by healthcare professionals, including physicians and pharmacists. We suggest the black box warning must have a barcode or logo on the outer pack of the drug if it has a black box warning, as this will give an alert for serious side effects or interactions for patient safety and provide an alert for physicians, patients, and pharmacists. Also, we suggest the warning black box drugs must be registered as notes in prescriptions and medical records at hospitals and in pharmacist and patient files, as this enhances health outcomes and avoids serious side effects for patient safety.

Fairness, especially group fairness, is an important consideration in the context of machine learning systems. The most commonly adopted group fairness-enhancing techniques are in-processing methods that rely on a mixture of a fairness objective (e.g., demographic parity) and a task-specific objective (e.g., cross-entropy) during the training process. However, when data arrives in an online fashion -- one instance at a time -- optimizing such fairness objectives poses several challenges. In particular, group fairness objectives are defined using expectations of predictions across different demographic groups. In the online setting, where the algorithm has access to a single instance at a time, estimating the group fairness objective requires additional storage and significantly more computation (e.g., forward/backward passes) than the task-specific objective at every time step. In this paper, we propose Aranyani, an ensemble of oblique decision trees, to make fair decisions in online settings. The hierarchical tree structure of Aranyani enables parameter isolation and allows us to efficiently compute the fairness gradients using aggregate statistics of previous decisions, eliminating the need for additional storage and forward/backward passes. We also present an efficient framework to train Aranyani and theoretically analyze several of its properties. We conduct empirical evaluations on 5 publicly available benchmarks (including vision and language datasets) to show that Aranyani achieves a better accuracy-fairness trade-off compared to baseline approaches.

BACKGROUND: Congenital leukemia is an exceptionally rare condition, with an incidence of 1 to 5 cases per million live births. Despite its rarity, the disease is clinically significant due to its severe manifestations and the need for tailored diagnostic and therapeutic strategies. Clinical features typically include hepatosplenomegaly, thrombocytopenia, and infiltrative cutaneous nodules. Recent advancements have identified the MLL gene as a critical player in leukemogenesis. A thorough understanding of the genetic and clinical aspects of congenital leukemia is essential for optimal patient management and genetic counseling. CASE PRESENTATION: A 1-day-old male newborn, the first child conceived through in vitro fertilization (IVF) for this family, presented with neonatal tachypnea, hepatosplenomegaly, and thrombocytopenia. Genetic analysis revealed a chromosomal translocation involving the MLL gene. His twin brother, conceived simultaneously but developing independently, showed no signs of leukemia. The therapeutic approach included oxygen therapy, platelet transfusions, and supportive care. Follow-up assessments indicated significant improvement in hepatosplenomegaly and hematological indices. CONCLUSION: This case highlights the clinical importance of congenital leukemia, particularly in the context of IVF-conceived pregnancies. The observed genetic discordance between the twins suggests the involvement of somatic mutations, chromosomal mosaicism, and epigenetic modifications. Recommendations for reducing the risk of genetic disorders in IVF-conceived pregnancies include preimplantation genetic testing, optimization of culture conditions, parental genetic screening, and ongoing research and education. This case represents the first documented instance worldwide of congenital leukemia occurring in one of the separate twins conceived through in vitro fertilization (IVF).

Purpose: This manuscript explores the potential of dual glucagon-like peptide 1 (GLP-1) agonists combined with degludec basal insulin as a treatment approach for early type 1 diabetes. The study aims to evaluate the efficacy and mechanistic impact of semaglutide, a GLP-1 agonist, on newly diagnosed type 1 diabetes patients. Methods: A retrospective analysis was conducted to assess the effects of semaglutide on individuals with early type 1 diabetes. The analysis focused on the elimination of prandial and basal insulin, changes in C-peptide levels, and overall glycemic control. The study also examined the potential for GLP-1 agonists to protect residual beta cells, stimulate cell proliferation, and reprogram liver cells into insulin-producing cells. Additionally, the modification of GLP-1 agonists with albumin ligands to extend their half-life and enhance their anti-diabetic effects was investigated. Results: The findings demonstrate the elimination of both prandial and basal insulin requirements, an increase in C-peptide levels, and improved glycemic control among the patients. Despite the positive outcomes, the study’s retrospective nature and absence of a control group highlight the necessity for larger, prospective trials. Conclusion: GLP-1 agonists show considerable potential in the management of type 1 diabetes by protecting residual beta cells, promoting cell proliferation, and reprogramming hepatic cells. The integration of modified GLP-1 agonists with albumin ligands could further enhance these effects. The manuscript underscores the need for continued research to fully explore this therapeutic approach. The proposed treatment strategy, which combines the autoimmune hypothesis, the proliferative effects of GLP-1, and albumin ligand modifications, aims to restore beta cell mass and function, thereby improving the quality of life for individuals with type 1 diabetes. Clinical trials are planned for 2024 under the registration ‹Amr Ahmed, Maher M. Akl, Semaglutide GLP1 Agonists with Degludec Basal-bolus Insulin in Early Type 1 Diabetes to Basalbolus› (ClinicalTrials.gov Identifier NCT06057077).

This manuscript delves into the intricate landscape of bladder cancer, highlighting the challenges of early detection and the nuanced considerations involved in chemotherapy decision-making based on patient-specific factors. The Maher Akl's groundbreaking Glucose Mutation Theory, presenting Glucosodiene as a promising breakthrough. The case presentation meticulously details the journey of a 72-year-old woman diagnosed with high-grade papillary urothelial carcinoma, spanning from initial diagnosis to postoperative outcomes. In the face of Bacillus Calmette-Guérin (BCG) therapy failure, Glucosodiene emerges as a safe and reliable alternative, offering a novel and effective treatment avenue, especially in cases where traditional chemotherapy is contraindicated. This approach advocates for Glucosodiene as a primary therapeutic option after the shortcomings of conventional treatments, instilling confidence in its efficacy and safety, particularly for cases deemed ineligible for chemotherapy.