Clinical and temporal characterization of COVID-19 subgroups using patient vector embeddings of electronic health records

Assessing the Quality of YouTube's Incontinence Information after Cancer Surgery: An Innovative Graphical Analysis

BACKGROUND

Prostate and colorectum cancers rank among the most common cancers, and incontinence is a significant postsurgical issue affecting the physical and psychological well-being of cancer survivors. Social media, particularly YouTube, has emerged as a vital source of health information. While YouTube offers valuable content, users must exercise caution due to potential misinformation.

OBJECTIVE

This study aims to assess the quality of publicly available YouTube videos related to incontinence after pelvic cancer surgery.

METHODS

A search on YouTube related to "Incontinence after cancer surgery" was performed, and 108 videos were analyzed. Multiple quality assessment tools (DISCERN, GQS, JAMA, PEMAT, and MQ-VET) and statistical analyses (descriptive statistics and intercorrelation tests) were used to evaluate the characteristics and popularity, educational value, quality, and reliability of these videos, relying on novel graphical representation techniques such as Sankey and Chord diagrams.

RESULTS

Strong positive correlations were found among quality rating scales, emphasizing agreement. The performed graphical analysis reinforced the reliability and validity of quality assessments.

CONCLUSIONS

This study found strong correlations among five quality scales, suggesting their effectiveness in assessing health information quality. The evaluation of YouTube videos consistently revealed "high" quality content. Considering the source is mandatory when assessing quality, healthcare and academic institutions are reliable sources. Caution is advised with ad-containing videos. Future research should focus on policy improvements and tools to aid patients in finding high-quality health content.

Applying unsupervised machine learning approaches to characterize autologous breast reconstruction patient subgroups: an NSQIP analysis of 14,274 patients

BACKGROUND

Autologous breast reconstruction is composed of diverse techniques and results in a variety of outcome trajectories. We propose employing an unsupervised machine learning method to characterize such heterogeneous patterns in large-scale datasets.

METHODS

A retrospective cohort study of autologous breast reconstruction patients was conducted through the National Surgical Quality Improvement Program database. Patient characteristics, intraoperative variables, and occurrences of acute postoperative complications were collected. The cohort was classified into patient subgroups via the K-means clustering algorithm, a similarity-based unsupervised learning approach. The characteristics of each cluster were compared for differences from the complementary sample (p < 2 ×10-4) and validated with a test set.

RESULTS

A total of 14,274 female patients were included in the final study cohort. Clustering identified seven optimal subgroups, ordered by increasing rate of postoperative complication. Cluster 1 (2027 patients) featured breast reconstruction with free flaps (50%) and latissimus dorsi flaps (40%). In addition to its low rate of complications (14%, p < 2 ×10-4), its patient population was younger and with lower comorbidities when compared with the whole cohort. In the other extreme, cluster 7 (1112 patients) almost exclusively featured breast reconstruction with free flaps (94%) and possessed the highest rates of unplanned reoperations, readmissions, and dehiscence (p < 2 ×10-4). The reoperation profile of cluster 3 was also significantly different from the general cohort and featured lower proportions of vascular repair procedures (p < 8 ×10-4).

CONCLUSIONS

This study presents a novel, generalizable application of an unsupervised learning model to organize patient subgroups with associations between comorbidities, modality of breast reconstruction, and postoperative outcomes.