Making medication prescriptions in response to the patient's diagnosis is a challenging task. The number of pharmaceutical companies, their inventory of medicines, and the recommended dosage confront a doctor with the well-known problem of information and cognitive overload. To assist a medical practitioner in making informed decisions regarding a medical prescription to a patient, researchers have exploited electronic health records (EHRs) in automatically recommending medication. In recent years, medication recommendation using EHRs has been a salient research direction, which has attracted researchers to apply various deep learning (DL) models to the EHRs of patients in recommending prescriptions. Yet, in the absence of a holistic survey article, it needs a lot of effort and time to study these publications in order to understand the current state of research and identify the best-performing models along with the trends and challenges. To fill this research gap, this survey reports on state-of-the-art DL-based medication recommendation methods. It reviews the classification of DL-based medication recommendation (MR) models, compares their performance, and the unavoidable issues they face. It reports on the most common datasets and metrics used in evaluating MR models. The findings of this study have implications for researchers interested in MR models.
Background and Objectives: Hereditary multiple exostoses (HME) is a disease characterized by cartilage-capped bony protuberances at the site of growth plates of long bones. Functional mutations in the exostosin genes (EXT1 and EXT2) are reported to affect the hedgehog signalling pathways leading to multiple enchondromatosis. However, the exact role of each EXT protein in the regulation of heparan sulphate (HS) chain elongation is still an enigma. In this study, a Pakistani family with HME is investigated to find out the genetic basis of the disease. Materials and Methods: Genotyping of eight members of the family by amplifying microsatellite markers, tightly linked to the EXT1 and EXT2 genes. Results: The study revealed linkage of the HME family to the EXT1 locus 8q24.1. Sanger sequencing identified a heterozygous deletion (c.247Cdel) in exon 1 of EXT1, segregating with the disease phenotype in the family. In silico analysis predicted a shift in the frame causing an early stop codon (p.R83GfsX52). The predicted dwarf protein constituting 134 amino acids was functionally aberrant with a complete loss of the catalytic domain at the C-terminus. Interestingly, an alternative open reading frame 3 (ORF3) caused by the frame shift is predicted to encode a protein sequence, identical to the wild type and containing the catalytic domain, but lacking the first 100 amino acids of the wild-type EXT1 protein. Conclusion: Consequently, haploinsufficiency could be the cause of HME in the investigated family as the mutated copy of EXT1 is ineffective for EXT-1/2 complex formation. The predicted ORF3 protein could be of great significance in understanding several aspects of HME pathogenesis.
Primary cholangiocarcinoma (malignancy of the bile ducts) is potentially a treatable malignancy via surgery. It presents with a derangement in the liver function blood test results, which results in raised bilirubin. It may also be accompanied by the complaint of itching of skin, dark urine, and light color stool. Bile duct metastasis from primary colorectal cancer, although a very rare condition presents with similar symptomatology and blood test results. Immunohistochemistry staining of tissue biopsy has significantly improved differentiation, detection, and hence plan management of both malignancies (cholangiocarcinoma and bile duct metastasis from primary colorectal cancer). The role of endoscopic retrograde cholangiopancreatography (ERCP) is important with the insertion of a stent towards symptom relief when palliative management is indicated in patients with an incurable disease.
17 November 2022 in Ibnosina Journal of Medicine and Biomedical Sciences
World Health Organization proposes implementing the WHO STEP smart strategy for tracking, to continuously monitoring noncommunicable diseases (NCDs) and associated risk factors (STEPS). Objective: To analyze the prevalence and contributing factors of NCD risk factors in Pakistan, a country with a low income where NCDs are responsible for 66 percent (or two-thirds) of fatalities annually. Methods: Total 675 eligible participants, ages 15 to 60, were selected by a multistage sampling approach from February 2020 and May 2021 for STEPS survey (version of instrument 3.2). Assessment of behavioral and physiological risk variables was part of the data collection process. SPSS version 25 was used for data analysis. Results: Intake of lesser than five portions of vegetables and fruits every day may be the greatest common risk factor. Of all participants, 341 (50.51%) were smoking currently, 513 (76%) were not physically active. A total of 251 (37.18%) individuals have elevated blood pressure whilst 245 (36.29%) individuals are affected by sugar. The cholesterol level is high in 215 (31.85%) participants out of 675. Each individual had an average of two risk factors for NCDs (2.04, 95percent [Confidence interval]: 2.02-2.08). Conclusions: Numerous risk factors for NCDs are present in a significant section of the Pakistani community. With the help of a multi-sectoral national coordination system, this observed data must be employed to promote and track specific NCD management and prevention programs for cities across Pakistan.