Tabular data is a common format for storing information in rows and columns, representing data entries and their associated features. Existing deep learning models designed for tabular data often rely on uni-dimensional feature selection mechanisms, which fail to adequately capture contextual dependencies of feature importance. This can result in missed crucial interactions, bias towards high-impact features, and limited generalization in attention generalization. To overcome these challenges, this study introduces a novel Iterative Feature Exclusion module that enhances the estimation of feature importance. The proposed module iteratively excludes a single feature from the input data and computes attention scores that represent the impact of each feature on the prediction. By aggregating the attention scores across iterations, the module generates a refined representation of feature importance that captures both the global and local feature interactions. The effectiveness of the proposed module was evaluated on four public datasets, and the results consistently demonstrated its superior performance over state-of-the-art methods and baseline models in both feature ranking and classification tasks.

The code is publicly available at https://github.com/abaraka2020/Iterative-Feature-Exclusion-Ranking-Module and https://github.com/mohalim/IFENet

Background: Gastrointestinal stromal tumors (GISTs) are the major mesenchymal tumors of the gastrointestinal 

tract (GIT), which progress due to the mutation in TKI and PDGFR-α receptors. It shows asymptomatic conditions 

however, become malignant due its large size and produce abdominal pain. Several surgical managements have 

been applied to diagnose patients with GIST, but these increase the risk of tumor recurrent rate. Therefore, 

adjuvant therapies contribute to preventing this challenge with several side effects and post-operative 

complications. A 42-year-old male patient from Libya with acute abdomen and GIST in small bowel (Jejunum) 

was diagnosed with complete excision besides lymph node dissection..Aim: To investigate numerous case reports 

through literature analysis regarding GIST along with a case presentation of the patient. Method: This research 

employed a literature-based study design. Previous case reports were screened from various online research 

platforms, including the National Center for Biotechnology Information (NCBI) and PubMed. A total of 37 cases 

were included through the search phrase "Case report on Gastrointestinal stromal tumor” .. Results: The findings 

revealed patient mean age (56.35) with most of the small intestine GIST (6.42 cm) site of origin, antibodies 

expression, and maximum mitotic cell counts.. Conclusion: The cases reported a risk of recurrence among GIST 

patients due to high mitotic counts and tumor size. Future researchers should address immune markers to reduce 

tumor recurrence among GIST patients. Stakeholders should implement proper treatment procedures with 

effective diagnostic methods. 

Keywords: Gastro-Intestinal Stromal Tumors (GISTs), Perforation, Imatinib, Acute abdomen, Tyrosine Kinase 

Inhibitor (TKI), Small Bowe

poultry industry in Libya has recently experienced significant growth, which is critical for 

improving food security. However, Toxoplasma gondii poses a threat to this industry because it infects warm-blooded 

animals, including poultry. Chickens, particularly free-range chickens, may serve as an important reservoir for T. 

gondii, yet there are no reports regarding the prevalence of this disease in free-range chickens in Libya. Understanding 

the seroprevalence of Toxoplasma in poultry is crucial for mitigating these risks and ensuring the health of the poultry 

industry.

Aim: This study examines the occurrence of T. gondii infection in poultry and evaluates its impact on public health in 

Libya.

Methods: A total of 315 free-range chickens were procured, with blood serum samples collected from the Al-Marj 

(n = 123), El-Bayda (n = 104), and Derna (n = 88) districts. The sera were analyzed using the indirect enzyme-linked 

immunosorbent assay. The chi-square test was used to evaluate a significant correlation between the seroprevalence of 

T. gondii and area, sex, and age.

Results: The overall seroprevalence of T. gondii antibodies was 23.8%. Among the districts, the highest prevalence 

was recorded in Al-Marj (26.8%), followed by El-Bayda (25%) and Derna (18.2%). Male chickens had a higher 

prevalence rate (26%; 25/96) than females (22.8%; 50/219), although no statistical difference was found between the 

genders. Interestingly, older chickens had a significantly higher seroprevalence (39.2%; 31/79) than younger chickens 

(28.6%; 24/84).

Conclusion: In Libya, free-range chickens are a potential source of Toxoplasma infection, with older chickens having 

a significantly higher seroprevalence than younger chickens. These free-range chickens could pose an infection risk 

to humans.

Background: Oral candidiasis is a common opportunistic infection, but its epidemiology varies significantly by

region. Recent data on the prevalence, causative species, and risk factors for oral fungal infections in Libya,

particularly in the Gharyan area, is scarce. Objective: This study aimed to determine the incidence, distribution

of Candida species, and associated demographic and health-related risk factors for oral candidiasis in Gharyan,

Libya. Methods: A cross-sectional study was conducted from February to April 2025 with 140 participants. Oral

swabs were collected following a clinical examination and structured interview. Samples were cultured on

Sabouraud Dextrose Agar, and Candida species were identified using chromogenic agar. Statistical analysis was

performed using SPSS version 27, employing Chi-square tests and ANOVA. Results: The incidence of oral

candidiasis was 47.9% (67/140). Candida albicans was the predominant species (85.1%), followed by C.

glabrata (6.0%), C. krusei (3.0%), C. parapsilosis (3.0%), and C. tropicalis (3.0%). Among the analyzed risk

factors (age, gender, diabetes, hypertension, smoking, dental prostheses, oral hygiene, and pregnancy), only

pregnancy showed a statistically significant association with infection (χ² = 4.486, *p* = 0.034). All pregnant

women in the sample were in the infected group. Conclusion: The study reveals a high incidence of oral

candidiasis in Gharyan, predominantly caused by C. albicans. Pregnancy was identified as a significant risk factor,

underscoring the need for targeted oral healthcare and screening for pregnant women. The lack of association with

other classic risk factors warrants further investigation with larger samples

In this study, we numerically investigate two significant medical models, Ebola Viral Disease (EVD) and Blood Ethanol Concentration (BEC) models-both formulated using Caputo-fractional derivatives. We develop and apply the Modified Fractional Euler Method (MFEM) for their solution, with a specific focus on error analysis. Comparative studies with the classical Runge-Kutta fourth-order method (RK4M) demonstrate that MFEM provides a computationally efficient and accurate alternative for solving such systems. The major features of the given procedure are its ease of application to this type of problem and other systems in various fields, in addition to the absence of numerical errors accumulating. Finally, we can control the increase in the convergence rate and the stability of the simulation process. The convergence examination and error estimation for the suggested scheme are also included. The importance of this study also lies in its contribution to our understanding of the dynamics of these two models in their fractional form. In addition, those numerical investigations demonstrate how control parameters affect specific components within these models.

Background In arid regions such as Jennawen, Libya, groundwater is the primary source for drinking and irrigation, yet it is highly vulnerable to geogenic and anthropogenic contamination. This study aimed to assess the chemical and microbiological quality of 26 groundwater samples from rural and remote, including desalinated water. Material and Methods Physico-chemical parameters (e.g. pH, salinity, major ions) and microbiological indicators (total coliforms, E. coli, total bacterial counts) were analyzed according to established standards. Results revealed that most untreated groundwater samples exceeded the permissible limits for salinity, hardness, and sulfate. Furthermore, microbial bacterial contamination was found at various locations. While some RO desalinated water was free from microbiological contaminants, it was severely deficient in essential minerals. Conclusion The groundwater in the investigated area requires proper attention. Research and management strategies are critically needed to address both chemical and microbial contamination while maintaining an optimal mineral balance in potable water.

The potential association between ABO/Rh blood groups and diabetes mellitus (DM) susceptibility remains a topic of global epidemiological interest, with population-specific findings. This systematic review aims to synthesize and analyze all available evidence on the distribution of ABO and Rh blood groups among diabetic patients across all cities in Libya to determine any consistent national pattern or association. A systematic search was conducted for studies published between 2010 and 2024. Electronic databases (PubMed, Google Scholar, Scopus) and Libyan journal archives were searched. Keywords included "ABO blood group," "Rhesus factor," "diabetes mellitus," "Libya," and specific city names. Observational studies reporting ABO/Rh frequencies in Libyan diabetic patients and controls were included. Data on study characteristics, blood group distribution, and odds ratios were extracted. Four studies from four major Libyan cities (Tripoli, Benghazi, Zliten, Zintan) met the inclusion criteria, encompassing 2,819 participants (1,919 diabetic patients and 900 controls). The pooled prevalence of blood group O was highest in both cases and controls. A meta-analysis of the pooled data revealed a significant association between blood group B and DM (OR: 1.52, 95% CI: 1.24-1.86, p < 0.001). Studies from Tripoli and Benghazi individually reported this significant association, while studies from Zliten and Zintan did not. No significant association was found between Rh factor and DM risk (OR: 1.08, 95% CI: 0.82-1.42, p = 0.59). This first systematic review from Libya indicates a potential national-level association between blood group B and an increased risk of diabetes mellitus. The lack of significance in two smaller studies highlights the need for larger, standardized, multi-center national research to confirm this finding and explore underlying genetic and environmental modifiers. The ABO blood group could be considered a modest genetic risk marker in the Libyan population.