Comparing gastrointestinal stromal tumor outcomes between geriatric and non-geriatric patients: A population-based analysis.

BACKGROUND: Gastrointestinal Stromal Tumors (GISTs) are the most common mesenchymal tumors of the GI tract. SEER is an extensive cancer database which proves useful in analyzing population trends. This analysis investigated GIST outcomes between geriatric & non-geriatric patients. METHODS: SEER*STAT 8.4.0.1 was used to extract relevant GIST data from 2000 to 2019. Geriatric age was defined as ≥70 years. Variables included age, sex, surgery, cancer-specific death, and overall survival. Statistical tests included univariate analysis using KM survival estimate (95% confidence interval) to calculate 5-year survival (5YS). Log-Rank tests determined statistical significance. Multivariable Cox's PH regression estimated the geriatric hazard death ratio adjusted for sex, stage, and surgery. RESULTS: The number of patients included was 13,579, yielding overall 5YS of 68.6% (95% CI 67.7-69.5). Cancer-specific death was 39.11% in 2000 & 3.33% in 2019. Non-geriatric & geriatric patient data yielded 5YS of 77.4% (76.4%-78.3%) and 53.3% (51.7%-54.8%) respectively (p < 0.0001). For no surgery/surgery, younger patient data yielded 5YS of 48.7% (45.8%-51.4%) and 83.7% (82.7%-84.7%) respectively (p < 0.0001); geriatric data yielded 5YS of 29.3% (26.5%-32.1%) and 62.8% (60.8%-64.6%) respectively (p < 0.0001). Multivariable analysis yielded a geriatric hazard death of 2.56 (2.42-2.70) (p < 0.0001). CONCLUSIONS: Cancer-specific death decreased since 2000, indicating an improvement in survival & treatment methods. Observed lower survival rates overall in the geriatric group. Surgery appeared to enhance survival rates in both groups, suggesting that surgery is an important factor in GIST survival regardless of age. Large prospective studies will help define clinical management for geriatric patients.

Patient Experience Performance at a Primary Cancer Center Versus Affiliated Community Facilities.

Purpose: Patient experience tools are used throughout health care to evaluate physician and departmental performance. In radiation medicine, these tools are important in evaluating patient-specific metrics throughout their care journey. This study compared patient experience outcomes from a central tertiary cancer program with network clinics in a health care network. Methods and Materials: Radiation medicine patient experience surveys (Press Ganey, LLC) were collected from a central facility and 5 network locations from January 2017 through June 2021. Surveys were distributed to patients after treatment completion. The study cohort was divided into the central facility and satellites. Questions were converted to a 0 to 100 scale from the Likert scale (1-5). To compare scores between site types, 2-way analysis of variance tests for the significance of sites adjusted for years of operations and adjustments for multiple comparisons (Dunnett's test) were completed on each question. Results: The number of consecutively returned surveys analyzed was 3777; a response rate of 33.3% was observed. The central site conducted 117,583 linear accelerator, 1425 Gamma Knife, 273 stereotactic radiosurgery, and 830 stereotactic body radiation therapy procedures. All satellites combined conducted 76,788 linear accelerator, 131 Gamma Knife, 95 stereotactic radiosurgery, and 355 stereotactic body radiation therapy procedures. The central facility fared better than the satellites on "Convenience of parking" (95.9 vs 87.9; P = .0001) but worse in other domains of care. Conclusions: All sites yielded exemplary patient experience rates. Community clinics scored higher than the main campus. The higher scores at the network sites require a deeper analysis of factors influencing the central facility, as the survey did not account for varying patient volumes and disparities in care complexity across sites. Attributes to satellites include lower patient volumes and easily navigable layouts. These results counter the impression that increased resources at the main campus create a better patient experience relative to network clinics and suggest that high-volume tertiary facilities will require unique initiatives to improve the patient experience.

Decoding Imagined Speech and Computer Control using Brain Waves.

BACKGROUND: In this work, we explore the possibility of decoding Imagined Speech (IS) brain waves using machine learning techniques. APPROACH: We design two finite state machines to create an interface for controlling a computer system using an IS-based brain-computer interface. To decode IS signals, we propose a covariance matrix of Electroencephalogram channels as input features, covariance matrices projection to tangent space for obtaining vectors from matrices, principal component analysis for dimension reduction of vectors, an artificial neural network (ANN) as a classification model, and bootstrap aggregation for creating an ensemble of ANN models. RESULT: Based on these findings, we are first to use an IS-based system to operate a computer and obtain an information transfer rate of 21-bits-per-minute. The proposed approach can decode the IS signal with a mean classification accuracy of 85% on classifying one long vs. short word. Our proposed approach can also differentiate between IS and rest state brain signals with a mean classification accuracy of 94%. COMPARISON: After comparison, we show that our approach performs equivalent to the state-of-the-art approach (SOTA) on decoding long vs. short word classification task. We also show that the proposed method outperforms SOTA significantly on decoding three short words and vowels with an average margin of 11% and 9%, respectively. CONCLUSION: These results show that the proposed approach can decode a wide variety of IS signals and is practically applicable in a real-time environment.