The efficacy of machine learning models in lung cancer risk prediction with explainability.

Among many types of cancers, to date, lung cancer remains one of the deadliest cancers around the world. Many researchers, scientists, doctors, and people from other fields continuously contribute to this subject regarding early prediction and diagnosis. One of the significant problems in prediction is the black-box nature of machine learning models. Though the detection rate is comparatively satisfactory, people have yet to learn how a model came to that decision, causing trust issues among patients and healthcare workers. This work uses multiple machine learning models on a numerical dataset of lung cancer-relevant parameters and compares performance and accuracy. After comparison, each model has been explained using different methods. The main contribution of this research is to give logical explanations of why the model reached a particular decision to achieve trust. This research has also been compared with a previous study that worked with a similar dataset and took expert opinions regarding their proposed model. We also showed that our research achieved better results than their proposed model and specialist opinion using hyperparameter tuning, having an improved accuracy of almost 100% in all four models.

A Systematic Review to Evaluate the Barriers to Breast Cancer Screening in Women with Disability.

Background: Breast cancer (BC) is one of the leading causes of mortality worldwide. There are observed disparities in patients with disability as compared to those without disability, which leads to poor BC screening attendance, thereby worsening disease management. Aim: The aim of this systematic review is to investigate if there are disparities in screening rates in women with disability as compared to those without disability, as well as the different factors that pose barriers to patients with disability for enrolment in BC screening programs. Method: Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we systematically reviewed published articles between 2008 and 2023, which assessed different factors that contributed to poor attendance in BC screening programs held across different countries. Detailed study characteristics were obtained, and methodological quality assessment was performed on the individual studies included in this review. Result: A total of fifty-three articles were identified as eligible studies based on the pre-defined inclusion and exclusion criteria. These included 7,252,913 patients diagnosed with BC (913,902 patients with disability/6,339,011 patients without disability). The results revealed there are demographic, clinical, financial, and service-related barriers that contributed to lower screening rates in disabled patients as compared to non-disabled. Patient age is the most common factor, with the highest effect observed for 80 years (vs. 30-44 years) [odds ratio (OR) = 13.93 (95% confidence interval (CI) = 8.27-23.47), p < 0.0001], followed by race/ethnicity for Hispanic (vs. non-Hispanic white) [OR = 9.5 (95%CI = 1.0-91.9), p < 0.05]. Additionally, patients with multiple disabilities had the highest rate of dropouts [OR = 27.4 (95%CI = 21.5-33.3)]. Other factors like education, income, marital status, and insurance coverage were essential barriers in screening programs. Conclusions: This study presents a holistic view of all barriers to poor BC screening attendance in disabled patients, thereby exacerbating health inequalities. A standardized approach to overcome the identified barriers and the need for a tailored guideline, especially for disability groups, is inevitable.

Optimizing CT Abdomen-Pelvis Scan Radiation Dose: Examining the Role of Body Metrics (Waist Circumference, Hip Circumference, Abdominal Fat, and Body Mass Index) in Dose Efficiency.

Objective: This study investigates the correlation between patient body metrics and radiation dose in abdominopelvic CT scans, aiming to identify significant predictors of radiation exposure. Methods: Employing a cross-sectional analysis of patient data, including BMI, abdominal fat, waist, abdomen, and hip circumference, we analyzed their relationship with the following dose metrics: the CTDIvol, DLP, and SSDE. Results: Results from the analysis of various body measurements revealed that BMI, abdominal fat, and waist circumference are strongly correlated with increased radiation doses. Notably, the SSDE, as a more patient-centric dose metric, showed significant positive correlations, especially with waist circumference, suggesting its potential as a key predictor for optimizing radiation doses. Conclusions: The findings suggest that incorporating patient-specific body metrics into CT dosimetry could enhance personalized care and radiation safety. Conclusively, this study highlights the necessity for tailored imaging protocols based on individual body metrics to optimize radiation exposure, encouraging further research into predictive models and the integration of these metrics into clinical practice for improved patient management.

Scintigraphy has the potential to replace thyroid stimulating hormone and ultrasonography in hyperthyroidism diagnosis.

The value of thyroid scintigraphy in hyperthyroidism diagnosis has long been the subject of debate. Unresolved issue is whether scintigraphy should be performed routinely, selectively, or for all hyperthyroidism patients. So, this study is concerned with the evaluation of thyroid scintigraphy for identifying hyperthyroidism in comparison with thyroid stimulating hormone (TSH) and ultrasound. This is cross sectional study including convenient patients sample (n = 50, 15 males and 35 females) aged (20-50 years) with primary hyperthyroidism and were attending endocrine clinics at King Faisal Specialist Hospital and Research Centre. All patients performed clinical investigations (TSH, ultrasound and thyroid scintigraphy). Among these patients, 96%, 48/50, had positive findings for hyperthyroidism with thyroid SC (95% CI; 96.0-99.5%); 84%, 42/50, had positive findings for hyperthyroidism by US (95% CI; 70.9-92.8%); and 56%, 28/50, had positive findings for hyperthyroidism by TSH measurement (95% CI; 41.3.0-70.0%). There was very good agreement between scintigraphy diagnosis and ultrasonography (kappa score = 0.812 (P < 0.0001), 95% CI (0.77-0.85). In many cases, scintigraphy provides considerably more functioning and anatomic details than ultrasound. In conclusion, these findings bring forth practical aspects of thyroid scintigraphy utilization for hyperthyroidism. By combining functional and anatomical information in one step, scintigraphy provides non-invasive, simple, fast and cost effective hyperthyroidism diagnostic method and has the potential to replace TSH and ultrasonography in hyperthyroidism investigation.