Search

Background: Breast cancer is the most common malignancy in women worldwide posing greater health challenge. The poor survival rate from breast cancer in India may be attributed to lack of awareness, low rate of screening uptake and presentation of cases in advanced stage. Objectives: To assess the baseline knowledge and effectiveness of educational intervention in improving the knowledge regarding breast cancer. Materials and Methods: A quasi-experimental community-based health educational intervention study was performed among 100 women and 18 female primary healthcare providers residing in urban field practice area. Face-to-face interviews were performed using a questionnaire adapted from the Breast Cancer Awareness Measure (Breast-CAM) Version 2. Data were analyzed using SPSS version 25. To assess the difference between pre- and postintervention scores, Mc-Nemar’s test and paired sample t tests were applied. Association between socio-demographic variables and difference in mean scores was assessed by using ANOVA and independent t test. P value less than 0.05 was taken as statistically significant. Results: The mean pre-intervention score of knowledge items (21 including warning signs, risk factors of breast cancer) among 100 women participants was 7.19 ± 3.45 and postintervention mean score was 18.18 ± 2.42. The mean pre-intervention score among 18 primary healthcare providers was 10.89 ± 4.59 and postintervention score was 19.89 ± 1.28. There was significant improvement in mean scores. The increase in percentage of scores in all 21 items postintervention was significant (Mc-Nemar’s test). There was significant association between socio-demographic profile and increase in mean scores. The study identified various barriers for seeking medical care among women. Conclusion: The community-based educational intervention was effective in enhancing the knowledge regarding breast cancer among women.

Conventional healthcare systems are traditionally challenged by fragmented data, lack of predictive insights, and security concerns, which spouse their effectiveness and efficiency. This paper will cover these gaps by developing an integrated Smart Healthcare System leveraging the power of the Internet of Things and Artificial Intelligence processes. To that end, we have proposed a holistic model that integrates several advanced methodologies to help in enhanced disease prediction and patient monitoring, with data security and privacy protection. We further apply the Hybrid Machine Learning (ML) models specifically; Random Forest Classifier integrated with k-means clustering for the prediction of diseases. This will cluster patients according to their similarity in health characteristics and provide an accurate disease risk prediction with an accuracy of 85-90%. Accordingly, Long Short Term Memory (LSTM) networks will be used for deeper timestamp series analyses with the following input sets: predicted disease probabilities, time-stamped health monitoring data, and patient lifestyle information sets. This model is outstanding both in regard to forecasting disease progression and in detecting anomalous health events with less than a 5% false positive rate. For protection and integrity of the data, we will use an Ethereum blockchain framework with respective smart contracts. The approach will provide secure, immutable health data storage and controlled, traceable access in full compliance with the requirements of various data protection regulations, such as GDPR. What's more, differentially private computations on encrypted data samples are guaranteed by combining homomorphic encryption methods with differential privacy techniques. The former ensures that in any kind of data analysis, at the point of execution, individual patient privacy is maintained, while the latter ensures an accurate, aggregated health data insight for different scenarios. By incorporating these methods, a robust smart healthcare system would be developed, one which, other than the ability to predict and monitor the progression of a disease very precisely, was able to protect patients' data and respect privacy. The same work has far-reaching implications in achieving better patient outcomes through earlier interventions and provision of increased security to the data, apart from enhancing trust in digital solutions for healthcare. [ABSTRACT FROM AUTHOR]