Predicting stereotactic radiosurgery outcomes with multi-observer qualitative appearance labelling versus MRI radiomics.

Qualitative observer-based and quantitative radiomics-based analyses of T1w contrast-enhanced magnetic resonance imaging (T1w-CE MRI) have both been shown to predict the outcomes of brain metastasis (BM) stereotactic radiosurgery (SRS). Comparison of these methods and interpretation of radiomics-based machine learning (ML) models remains limited. To address this need, we collected a dataset of n = 123 BMs from 99 patients including 12 clinical features, 107 pre-treatment T1w-CE MRI radiomic features, and BM post-SRS progression scores. A previously published outcome model using SRS dose prescription and five-way BM qualitative appearance scoring was evaluated. We found high qualitative scoring interobserver variability across five observers that negatively impacted the model's risk stratification. Radiomics-based ML models trained to replicate the qualitative scoring did so with high accuracy (bootstrap-corrected AUC = 0.84-0.94), but risk stratification using these replicated qualitative scores remained poor. Radiomics-based ML models trained to directly predict post-SRS progression offered enhanced risk stratification (Kaplan-Meier rank-sum p = 0.0003) compared to using qualitative appearance. The qualitative appearance scoring enabled interpretation of the progression radiomics-based ML model, with necrotic BMs and a subset of heterogeneous BMs predicted as being at high-risk of post-SRS progression, in agreement with current radiobiological understanding. Our study's results show that while radiomics-based SRS outcome models out-perform qualitative appearance analysis, qualitative appearance still provides critical insight into ML model operation.

Dual-center validation of using magnetic resonance imaging radiomics to predict stereotactic radiosurgery outcomes.

Background: MRI radiomic features and machine learning have been used to predict brain metastasis (BM) stereotactic radiosurgery (SRS) outcomes. Previous studies used only single-center datasets, representing a significant barrier to clinical translation and further research. This study, therefore, presents the first dual-center validation of these techniques. Methods: SRS datasets were acquired from 2 centers (n = 123 BMs and n = 117 BMs). Each dataset contained 8 clinical features, 107 pretreatment T1w contrast-enhanced MRI radiomic features, and post-SRS BM progression endpoints determined from follow-up MRI. Random decision forest models were used with clinical and/or radiomic features to predict progression. 250 bootstrap repetitions were used for single-center experiments. Results: Training a model with one center's dataset and testing it with the other center's dataset required using a set of features important for outcome prediction at both centers, and achieved area under the receiver operating characteristic curve (AUC) values up to 0.70. A model training methodology developed using the first center's dataset was locked and externally validated with the second center's dataset, achieving a bootstrap-corrected AUC of 0.80. Lastly, models trained on pooled data from both centers offered balanced accuracy across centers with an overall bootstrap-corrected AUC of 0.78. Conclusions: Using the presented validated methodology, radiomic models trained at a single center can be used externally, though they must utilize features important across all centers. These models' accuracies are inferior to those of models trained using each individual center's data. Pooling data across centers shows accurate and balanced performance, though further validation is required.

Performance sensitivity analysis of brain metastasis stereotactic radiosurgery outcome prediction using MRI radiomics.

Recent studies have used T1w contrast-enhanced (T1w-CE) magnetic resonance imaging (MRI) radiomic features and machine learning to predict post-stereotactic radiosurgery (SRS) brain metastasis (BM) progression, but have not examined the effects of combining clinical and radiomic features, BM primary cancer, BM volume effects, and using multiple scanner models. To investigate these effects, a dataset of n = 123 BMs from 99 SRS patients with 12 clinical features, 107 pre-treatment T1w-CE radiomic features, and BM progression determined by follow-up MRI was used with a random decision forest model and 250 bootstrapped repetitions. Repeat experiments assessed the relative accuracy across primary cancer sites, BM volume groups, and scanner model pairings. Correction for accuracy imbalances across volume groups was investigated by removing volume-correlated features. We found that using clinical and radiomic features together produced the most accurate model with a bootstrap-corrected area under the receiver operating characteristic curve of 0.77. Accuracy also varied by primary cancer site, BM volume, and scanner model pairings. The effect of BM volume was eliminated by removing features at a volume-correlation coefficient threshold of 0.25. These results show that feature type, primary cancer, volume, and scanner model are all critical factors in the accuracy of radiomics-based prognostic models for BM SRS that must be characterised and controlled for before clinical translation.

Preventing, Identifying, and Treating Prescription Drug Misuse Among Active-Duty Service Members.

Prescription drug misuse (PDM) is of critical concern for the military because of its potential impact on military readiness, the health and well-being of military personnel, and associated health care costs. The purpose of this study is to summarize insights gleaned from a series of activities that the RAND Corporation undertook for the Deputy Assistant Secretary of Defense for Readiness to address this important health and military readiness issue. The authors completed a review of U.S. Department of Defense policies and a comprehensive literature review of clinical guidelines and the empirical literature on the prevention and treatment of PDM and conducted individual face-to-face interviews with 66 health and behavioral health care providers at nine medical treatment facilities across three regions within the contiguous United States to identify best practices in the prevention, identification, and treatment of PDM and the extent to which those practices are known and followed. The study also presents the framework of an analytic tool that, once informed by data available to the military but not available to the authors, can assist the military in predicting future trends in PDM based on current demographics of active-duty service members and rates of injury and prescribing of prescription drugs. The findings from this work led the authors to formulate a set of key insights that they believe might improve the rapid identification and treatment of service members dealing with PDM, thereby improving future force readiness.

Bench-scale biodegradation tests to assess natural attenuation potential of 1,4-dioxane at three sites in California.

1,4-Dioxane (dioxane) is relatively recalcitrant to biodegradation, and its physicochemical properties preclude effective removal from contaminated groundwater by volatilization or adsorption. Through this microcosm study, we assessed the biodegradation potential of dioxane for three sites in California. Groundwater and sediment samples were collected at various locations at each site, including the presumed source zone, middle and leading edge of the plume. A total of 16 monitoring wells were sampled to prepare the microcosms. Biodegradation of dioxane was observed in 12 of 16 microcosms mimicking natural attenuation within 28 weeks. Rates varied from as high as 3,449 ± 459 µg/L/week in source-zone microcosms to a low of 0.3 ± 0.1 µg/L/week in microcosms with trace level of dioxane (<10 µg/L as initial concentration). The microcosms were spiked with (14)C-labeled dioxane to assess the fate of dioxane. Biological oxidizer-liquid scintillation analysis of bound residue infers that 14C-dioxane was assimilated into cell material only in microcosms exhibiting significant dioxane biodegradation. Mineralization was also observed per (14)CO2 recovery (up to 44% of the amount degraded in 28 weeks of incubation). Degradation and mineralization activity significantly decreased with increasing distance from the contaminant source area (p < 0.05), possibly due to less acclimation. Furthermore, both respiked and repeated microcosms prepared with source-zone samples from Site 1 confirmed relatively rapid dioxane degradation (i.e., 100 % removal by 20 weeks). These results show that indigenous microorganisms capable of degrading dioxane are present at these three sites, and suggest that monitored natural attenuation should be considered as a remedial response.

Gender difference in stroke case fatality: an integrated study of hospitalization and mortality.

BACKGROUND: Providing regional and state-specific prognosis factors for stroke patients has both clinical and public health importance. Results from previous studies of sex difference in stroke case fatality have been mixed. The current study links stroke hospitalizations to community-based mortality records to examine sex difference in stroke case fatality and associated prognosis factors. METHODS: Hospital discharge data and death certificate data from January 2005 to December 2009 in Nebraska were linked. Multivariable logistic regression was used to estimate sex differences in 30-day mortality, and the Cox proportional hazard model was used to predict overall survival. RESULTS: A total of 15,806 patients were included. Females were more likely to die during the 30 days after stroke hospitalization. However, there was no significant difference in overall survival in the multivariate analysis that controlled for age, comorbidity, and rehabilitation factors. Females were more likely to have comorbidities, such as atrial fibrillation, anemia, and heart failure, while males were more likely to have chronic kidney disease. In addition, males were more likely to receive rehabilitation services after stroke. CONCLUSIONS: Among persons hospitalized with a stroke in Nebraska between 2005 and 2009, the crude case fatality rate was 50% higher in women. However, after accounting for age and other variables, adjusted mortality rates were essentially the same for men and women.

Genitourinary complications of inherited epidermolysis bullosa: experience of the national epidermylosis bullosa registry and review of the literature.

PURPOSE: We determined the frequency with which 6 genitourinary tract complications (urethral meatal stenosis, urinary retention, bladder hypertrophy, hydronephrosis secondary to ureteral strictures, pyelonephritis and cystitis) occur in inherited epidermolysis bullosa (EB) in the American EB population. MATERIALS AND METHODS: Up to 16 years of longitudinal followup was done in 3,280 consecutively enrolled patients in the National EB Registry, a National Institutes of Health funded epidemiological study covering the entire continental United States. Data were stratified by major EB type and subtype. Frequencies of occurrence were determined for each of 6 genitourinary tract variables and stratified into 10 mutually exclusive, major EB subtypes. RESULTS: Urinary tract complications occurred in a minority of patients across all major EB subtypes with the highest frequency seen in Herlitz junctional EB (JEB-H). Urethral meatus stenosis was the most common complication, occurring in 11.6% and 8.0% of patients with JEB-H and Hallopeau-Siemens recessive dystrophic EB (RDEB), respectively. Urinary retention, hydronephrosis and bladder hypertrophy occurred in 9.3%, 7.0% and 4.6% of JEB-H cases, respectively. In contrast, pyelonephritis and cystitis were most often seen in the setting of generalized EB simplex (Koebner variant) and inversa RDEB. CONCLUSIONS: The urinary tract may be involved in any subtype of inherited EB, although these complications usually arise in patients with the most severe subtypes of junctional and recessive dystrophic disease. Chronic surveillance for the presence of genitourinary tract disease activity is warranted, especially in patients with JEB and RDEB, given the potential for longterm kidney injury if untreated.

Inherited epidermolysis bullosa and the risk of death from renal disease: experience of the National Epidermolysis Bullosa Registry.

BACKGROUND: Isolated case reports and limited clinical series suggested that renal failure may occur in some patients with inherited epidermolysis bullosa (EB). Reported causes have included poststreptococcal glomerulonephritis, secondary amyloidosis, and chronic mechanical obstruction. To date, no data exist that permit an accurate estimation of the risk for death from renal failure in patients with this disease. METHODS: Life table analyses were performed using 16 years of data from the National EB Registry, a federally funded longitudinal epidemiological study encompassing the entire continental United States. Data were stratified so that cumulative and conditional risks for death from renal failure could be estimated. Renal failure as cause of death was identified by death certificates and verified from hospital records and interviews with the patients' immediate families. RESULTS: The cumulative risk for death from renal failure among patients with Hallopeau-Siemens recessive dystrophic EB (RDEB-HS) was 12.3% by age 35 years. In addition, deaths from renal failure also were reported rarely in patients with other subtypes of generalized RDEB and in junctional EB (JEB). CONCLUSION: Renal failure is an important cause of death among adults with RDEB-HS, surpassed only by death from metastatic squamous cell carcinoma. It also rarely may occur in the setting of JEB and other subtypes of generalized RDEB. Given our data, medical surveillance for early renal involvement should become part of the routine evaluation of all adults with RDEB and JEB.

Eye involvement in inherited epidermolysis bullosa: experience of the National Epidermolysis Bullosa Registry.

PURPOSE: To determine the frequency of ocular manifestations in inherited epidermolysis bullosa (EB) within the continental United States and to define the estimated cumulative risks of developing nonscarring (blisters or erosions) and scarring corneal manifestations within each major EB subtype over time. DESIGN: Observational (cross-sectional and longitudinal). METHODS: Up to 16 years of longitudinal follow-up was conducted on 3,280 consecutively enrolled patients in the National EB Registry, an epidemiologic study funded by the National Institutes of Health. Data were stratified by major EB type and subtype. Frequencies of occurrence were determined for eight variables (corneal erosions or blistering; corneal scarring; symblepharons; blepharitis; ectropions; lacrimal duct obstruction; impaired vision; blindness) by contingency tables, and cumulative risks were generated by life table analysis technique. RESULTS: The most common ocular manifestations were corneal erosions and blisters. Frequencies mirrored relative severity of skin disease, with 74.10% of all patients with recessive dystrophic EB, Hallopeau-Siemens (RDEB-HS) and 47.50% of all patients with junctional EB, Herlitz (JEB-H) experiencing at least one episode. Lower frequencies were noted for corneal scarring. Symblepharons and ectropions were most commonly seen in inversa RDEB and JEB-H, respectively. Blindness was reported in 6.47% of RDEB-HS patients. The cumulative risks of nonscarring and scarring corneal lesions in JEB-H at age 5 are 83.18% and 27.08% and at age 25 are 83.18% and 72.22%. With time, the cumulative risk of each in RDEB-HS approached that reported in JEB-H patients. CONCLUSION: Ocular disease activity, particularly corneal, is common in some EB subtypes. Careful ophthalmologic examination should become an integral part of the management of all patients with inherited EB.

Relative extent of skin involvement in inherited epidermolysis bullosa (EB): composite regional anatomic diagrams based on the findings of the National EB Registry, 1986 to 2002.

BACKGROUND: Inherited epidermolysis bullosa (EB) encompasses 4 major types and at least 23 clinically distinctive phenotypes. Although considerable variability in cutaneous disease activity is known to exist within each, severity and anatomic distribution of skin lesions remain the major criteria used for subclassification. OBJECTIVE: We sought to generate accurate anatomic "density" diagrams depicting the relative extent and location of skin lesions within each major EB subtype. METHODS: Diagrams were created for each major EB type, on the basis of medical history and physical examination findings obtained from 1986 to 2002 from 3280 consecutive enrollees in the National EB Registry. RESULTS: An anatomic diagram was created for each of the major EB subtypes, representing a prototypic composite photograph of cutaneous disease activity. CONCLUSIONS: Marked variability exists in the extent of skin involvement within each major EB subtype. The use of these diagrams, generated from the world's largest cohort of patients with EB, should assist the clinician in more accurately subclassifying newly encountered patients.

Medical and dermatology dictionaries: an examination of unstructured definitions and a proposal for the future.

Medical dictionaries serve to describe and clarify the term set used by medical professionals. In this commentary, we analyze a representative set of skin disease definitions from 2 prominent medical dictionaries, Stedman's Medical Dictionary and Dorland's Illustrated Medical Dictionary. We find that there is an apparent lack of stylistic standards with regard to content and form. We advocate a new standard form for the definition of medical terminology, a standard to complement the easy-to-read yet unstructured style of the traditional dictionary entry. This new form offers a reproducible structure, paving the way for the development of a computer readable "dictionary" of medical terminology. Such a dictionary offers immediate update capability and a fundamental improvement in the ability to search for relationships between terms.