Revisiting the Risk Factors for Endometriosis: A Machine Learning Approach.

Endometriosis is a condition characterized by implants of endometrial tissues into extrauterine sites, mostly within the pelvic peritoneum. The prevalence of endometriosis is under-diagnosed and is estimated to account for 5-10% of all women of reproductive age. The goal of this study was to develop a model for endometriosis based on the UK-biobank (UKB) and re-assess the contribution of known risk factors to endometriosis. We partitioned the data into those diagnosed with endometriosis (5924; ICD-10: N80) and a control group (142,723). We included over 1000 variables from the UKB covering personal information about female health, lifestyle, self-reported data, genetic variants, and medical history prior to endometriosis diagnosis. We applied machine learning algorithms to train an endometriosis prediction model. The optimal prediction was achieved with the gradient boosting algorithms of CatBoost for the data-combined model with an area under the ROC curve (ROC-AUC) of 0.81. The same results were obtained for women from a mixed ethnicity population of the UKB (7112; ICD-10: N80). We discovered that, prior to being diagnosed with endometriosis, affected women had significantly more ICD-10 diagnoses than the average unaffected woman. We used SHAP, an explainable AI tool, to estimate the marginal impact of a feature, given all other features. The informative features ranked by SHAP values included irritable bowel syndrome (IBS) and the length of the menstrual cycle. We conclude that the rich population-based retrospective data from the UKB are valuable for developing unified machine learning endometriosis models despite the limitations of missing data, noisy medical input, and participant age. The informative features of the model may improve clinical utility for endometriosis diagnosis.

Cannabis and Pain Treatment-A Review of the Clinical Utility and a Practical Approach in Light of Uncertainty.

Over the past decade the phenomenon of cannabis as a legitimate form of treatment for pain has overwhelmed the medical community, especially in the field of pain. From a status of a schedule 1 substance having no currently accepted medical use and being considered to have high potential for abuse, its use has mushroomed to over 50,000 legal medical users per year in Israel alone. There appear to be many reasons behind this phenomenon-medical, sociological, and economical. Thus, what is cannabis? An abusive substance or a medication? Should it be incorporated into current biomedical practice, and how should it be administered? Finally, what is the evidence for the beneficial and detrimental effects of cannabis? This article reviews and discusses the current literature regarding the beneficial and the detrimental effects of medical cannabis in the treatment of pain. We further discuss the problems and challenges facing the medical community in this domain and offer a practical approach to deal with these challenges.

Insoles for prevention and treatment of back pain: a systematic review within the framework of the Cochrane Collaboration Back Review Group.

STUDY DESIGN: A systematic review of randomized controlled trials. OBJECTIVE: To determine the effectiveness of shoe insoles in the prevention and treatment of nonspecific back pain compared with placebo, no intervention, or other interventions. SUMMARY OF BACKGROUND DATA: There is lack of theoretical and clinical knowledge of the use of insoles for prevention or treatment of back pain. METHODS: We searched electronic databases from inception to October 2008. We reviewed reference lists in review articles, guidelines, and in the included trials; conducted citation tracking; and contacted individuals with expertise in this domain. One review author conducted the searches and blinded the retrieved references for authors, institution, and journal. Two review authors independently selected the relevant articles. Two different review authors independently assessed the methodological quality and clinical relevance and extracted the data from each trial using the criteria recommended by the Cochrane Back Review Group. RESULTS: Six randomized controlled trials met inclusion criteria: 3 examined prevention of back pain (2061 participants) and 3 examined mixed populations (256 participants) without being clear whether they were aimed at primary or secondary prevention or treatment. No treatment trials were found. There is strong evidence that the use of insoles does not prevent back pain. There is limited evidence that insoles alleviate back pain or adversely shift the pain to the lower extremities. CONCLUSION: There is strong evidence that insoles are not effective for the prevention of back pain. The current evidence on insoles as treatment for low back pain does not allow any conclusions.

Antibiotic prophylaxis for spontaneous bacterial peritonitis in cirrhotic patients with ascites, without gastro-intestinal bleeding.

BACKGROUND: Spontaneous bacterial peritonitis is frequent among cirrhotic patients, associated with significant morbidity and mortality. Selective intestinal decontamination employing antibiotics is a proposed prophylactic measure. While data regarding this modality among cirrhotic patients with gastrointestinal bleeding exist, there is insufficient data synthesis regarding cirrhotic patients with ascites and no gastrointestinal bleeding. OBJECTIVES: To assess whether antibiotic prophylaxis decreases spontaneous bacterial peritonitis and mortality among cirrhotic patients with ascites and no gastrointestinal bleeding. SEARCH STRATEGY: We identified relevant randomised trials by searching trial registries of The Cochrane Hepato-Biliary Group and The Cochrane Collaboration, medical literature search engines, and reviewing all literature we found on the topic until February 2009. SELECTION CRITERIA: We searched for randomised clinical trials assessing prophylactic treatment among adult cirrhotic patients with ascites and no gastrointestinal bleeding, comparing antibiotic therapy with no intervention, placebo, or with another antibiotic regimen. DATA COLLECTION AND ANALYSIS: Three independent authors searched for and collected the trials and extracted relevant data. Four other independent authors validated the findings and assessed them. The studies were assessed for design, patient and intervention characteristics, and quality. A meta-analysis was performed to estimate measures of association between antibiotic prophylaxis and spontaneous bacterial peritonitis or mortality. MAIN RESULTS: Nine trials were included in the review. Seven trials, comparing antibiotics to placebo or no treatment, were meta-analysed. Systematic bias in design or publication is suggested by trial results. The randomisation results suggest that the probability that true randomisation took place in all trials is very small and the report of most trials regarding design was poor. The proportion of participants with spontaneous bacterial peritonitis varied between the trials from 15% to 50%. The calculated relative risks (95% confidence interval) of spontaneous bacterial peritonitis and mortality among patients treated with antibiotics compared with no treatment/placebo were 0.20 (0.11 to 0.37) and 0.61 (0.43 to 0.87). There were very few reports of adverse events. AUTHORS' CONCLUSIONS: The pooled estimates suggest that antibiotic prophylaxis might be prudent among cirrhotic patients with ascites and no gastrointestinal bleeding. However, poor trial methodology and report coupled with findings suggesting systematic bias in publication and design reflect the fragility of these findings. Potential hazard to society and the patients themselves from resistant pathogens should be considered when promoting long-lasting antibiotic prophylaxis. It seems that recommending antibiotic prophylaxis is still far from being a substantiated prevention strategy. Trials of better design, well reported, and of longer follow-up are greatly needed.