Search

This essay presents quantitative capture and predictive modeling for one of the largest and longest running mass reading programs of the past two decades: One Book One Chicago (OBOC) sponsored by the Chicago Public Library (CPL). The Reading Chicago Reading project uses data associated with OBOC as a probe into city-scale library usage and, by extension, as a window onto contemporary reading behavior. The first half of the essay explains why CPL's OBOC program is conducive for modeling purposes, and the second half documents the creation of our models, their underlying data, and the results. Explores data from the Chicago Public Library's One Book One Chicago project to analyze and model library usage and reading behavior [ABSTRACT FROM AUTHOR]

Simple Summary: This research aimed to investigate if adding social determinants of health (SDOH) to predictive models improves major adverse cardiovascular events (MACE) predictions in breast cancer patients, as cardiovascular disease is their leading cause of death. ML models, incorporating SDOH, demographics, risk factors, tumor characteristics, and treatments, were developed and compared. The results showed that including SDOH enhanced ML model performance in forecasting MACEs within two years of breast cancer diagnosis, especially for non-Hispanic Black patients. These findings offer more accurate risk assessments and personalized care insights for breast cancer patients, while also guiding efforts toward achieving healthcare equity. Cardiovascular disease is the leading cause of mortality among breast cancer (BC) patients aged 50 and above. Machine Learning (ML) models are increasingly utilized as prediction tools, and recent evidence suggests that incorporating social determinants of health (SDOH) data can enhance its performance. This study included females ≥ 18 years diagnosed with BC at any stage. The outcomes were the diagnosis and time-to-event of major adverse cardiovascular events (MACEs) within two years following a cancer diagnosis. Covariates encompassed demographics, risk factors, individual and neighborhood-level SDOH, tumor characteristics, and BC treatment. Race-specific and race-agnostic Extreme Gradient Boosting ML models with and without SDOH data were developed and compared based on their C-index. Among 4309 patients, 11.4% experienced a 2-year MACE. The race-agnostic models exhibited a C-index of 0.78 (95% CI 0.76–0.79) and 0.81 (95% CI 0.80–0.82) without and with SDOH data, respectively. In non-Hispanic Black women (NHB; n = 765), models without and with SDOH data achieved a C-index of 0.74 (95% CI 0.72–0.76) and 0.75 (95% CI 0.73–0.78), respectively. Among non-Hispanic White women (n = 3321), models without and with SDOH data yielded a C-index of 0.79 (95% CI 0.77–0.80) and 0.79 (95% CI 0.77–0.80), respectively. In summary, including SDOH data improves the predictive performance of ML models in forecasting 2-year MACE among BC females, particularly within NHB. [ABSTRACT FROM AUTHOR]

PURPOSE: Develop a cancer-specific machine learning (ML) model that accurately predicts 30-day unplanned readmissions in patients with solid tumors. METHODS: The initial cohort included patients 18 years or older diagnosed with a solid tumor. Two distinct cohorts were generated: one with and one without detailed social determinants of health (SDOHs) data. For each cohort, data were temporally partitioned in 70% (training), 20% (validation), and 10% (testing). Tree-based ML models were developed and validated on each cohort. The metrics used to evaluate the model's performance were receiver operating characteristic curve (ROC), area under the ROC curve, precision, recall (R), accuracy, and area under the precision-recall curve. RESULTS: We included 13,717 patients in this study in two cohorts (5,059 without SDOH data and 8,658 with SDOH data). Unplanned 30-day readmission occurred in 21.3% of the cases overall. The five main non-SDOH factors most highly associated with an unplanned 30-day readmission (R, 0.74; IQR, 0.58-0.76) were: number of previous unplanned readmissions; higher Charlson comorbidity score; nonelective index admission; discharge to anywhere other than home, hospice, or nursing facility; and higher anion gap during the admission. Neighborhood crime index, neighborhood median home values, annual income, neighborhood median household income, and wealth index were the main five SDOH factors important for predicting a high risk for an unplanned hospital readmission (R, 0.66; IQR, 0.56-0.72). The models were not directly comparable. CONCLUSION: Key drivers of unplanned readmissions in patients with cancer are complex and involve both clinical factors and SDOH. We developed a cancer-specific ML model that with reasonable accuracy identified patients with cancer at high risk for an unplanned hospital readmission. New cancer-specific ML models accurately predict 30-day unplanned readmissions in patients with solid tumors. [ABSTRACT FROM AUTHOR]

Plant N concentration (PNC) has been commonly used to guide farmers in assessing maize (Zea mays L.) N status and making in-season N fertilization decisions. However, PNC varies based on the development stage. Therefore, a relationship between biomass and N concentration is needed (i.e., critical N dilution curve; CNDC) to better understand when plants are N deficient. A few CNDCs have been developed and used for plant N status diagnoses but have not been tested in the US Midwest. The objective of this study was to evaluate under highly diverse soil and weather conditions in the US Midwest the performance of CNDCs developed in France and China for assessing maize N status. Maize N rate response trials were conducted across eight US Midwest states over three years. This analysis utilized plant and soil measurements at V9 and VT development stages and final grain yield. Results showed that the French CNDC (y = 34.0x−0.37, where y is critical PNC, and x is aboveground biomass) was better with a 91% N status classification accuracy compared to only 62% with the Chinese CNDC (y = 36.5x−0.48). The N nutrition index (NNI), which is the quotient of the measured PNC and the calculated critical N concentration (Nc) based on the French CNDC was significantly related to soil nitrate-N content (R2 = 0.38–0.56). Relative grain yield on average reached a plateau at NNI values of 1.36 at V9 and 1.21 at VT but for individual sites ranging from 0.80 to 1.41 at V9 and from 0.62 to 1.75 at VT. The NNI threshold values or ranges optimal for crop biomass production may not be optimal for grain yield production. It is concluded that the CNDC developed in France is suitable as a general diagnostic tool for assessing maize N status in US Midwest. However, the threshold values of NNI for diagnosing maize N status and guiding N applications vary significantly across the region, making it challenging to guide specific on-farm N management. More studies are needed to determine how to effectively use CNDC to make in-season N recommendations in the US Midwest. [ABSTRACT FROM AUTHOR]

Improving corn (Zea mays L.) nitrogen (N) rate fertilizer recommendation tools can improve farmers' profits and mitigate N pollution. Numerous approaches have been tested to improve these tools, but to date improvements for predicting economically optimum N rate (EONR) have been modest. This work's objective was to use ensemble learning to improve our estimation of EONR (for a single at‐planting and split N application timing) by combining multiple corn N recommendation tools. The evaluation was conducted using 49 corn N response trials from eight states in the US Corn Belt and three growing seasons (2014–2016). Elastic net and decision tree approaches regressed EONR against three unique tools for each N application timing. Tools used in various combinations included a yield goal method, two soil nitrate tests (pre‐plant and late season), a computer simulation crop model (Maize‐N), and canopy reflectance sensing. Any combination of two or three N recommendation tools improved or maintained performance metrics (R2, root‐mean square error , and number of sites close to EONR). The best results for a single at‐planting recommendation occurred when combining the three at‐planting N recommendation tools (including interactions) with an elastic net regression model. This combined recommendation tool had a significant linear relationship with EONR (R2 = 0.46), an increase of 0.27 over the best tool evaluated alone. Combining multiple tools increased the implementation cost, but it did not reduce profitability and, sometimes, improved profitability. These results show tools can be combined to better match EONR, and thus could aid farmers in improving N management. Core Ideas: Individual corn N recommendation tools poorly estimated EONR (R2 ≤ 0.24) across many US Midwest states.Combining or "ensembling" two or more corn N recommendation tools improved EONR estimation (R2 ≤ 0.46).Both ensembling techniques evaluated proved effective at combining tools. [ABSTRACT FROM AUTHOR]

Background: Patients with certain autoimmune conditions are at a reduced risk of developing breast cancer compared to the general population. Despite this, little is known about outcomes in patients with breast cancer who have a concurrent autoimmune diagnosis. Methods: This study compared differences in outcomes between women with breast cancer who had or did not have an autoimmune diagnosis. The SEER‐Medicare databases (2007–2014) were used to identify patients with breast cancer and diagnosis codes were used to identify those with an autoimmune disorder. Results: The studied autoimmune diseases had a prevalence of 27% among the 137,324 patients with breast cancer. Autoimmune disease was associated with significantly longer overall survival (OS) and significantly lower cancer‐specific mortality (CSM) among stage IV breast cancer patients (p < 0.0001). After controlling for the effects of age, race, chronic kideny disease, chemotherapy, and radiation therapy autoimmune disease was still predictive of improved OS (HR: 1.45, 95% CI: 1.35–1.55, p < 0.0001) and CSM (HR: 1.40, 95% CI: 1.29–1.5, p < 0.0001). By contrast, in patients with stage I–III breast cancer, the presence of an autoimmune diagnosis was associated with a lower OS (p < 0.0001, p < 0.0001, and p = 0.026, respectively), compared to patients without autoimmune disease. Conclusions: We found a higher prevalence of rheumatoid arthritis, Crohn's disease, ulcerative colitis, and systemic lupus erythematosus in patients with breast cancer compared to age matched cohorts in the general population. The presence of an autoimmune diagnosis was associated with a lower OS in stages I–III breast cancer and improved OS and CSM in patients with stage IV disease. These results suggest that anti‐tumor immunity plays an important role in late stage breast cancer and could potentially be exploited to improve the effectiveness of immunotherapy. [ABSTRACT FROM AUTHOR]

Background: There is a male predominance of acute myeloid leukemia (AML) incidence, but survival data are conflicting. The objective of this study is to carry out a comprehensive analysis of sex differences in AML, and to investigate the impact of sex disparities in survival. Methods: The cohort included patients ≥18 years diagnosed with AML (2010–2022). Demographics, treatment patterns, treatment adverse events, and survival were analyzed. The population was described and compared by sex, and sex‐based risks and associations were obtained via Cox proportional‐hazards regression. Results: In total, 1020 AML patients were analyzed (57.4% males), with lower risk of death for females (aHR = 0.41, 95% CI 0.26–0.66). Among females, BMT (aHR = 0.51, 95% CI 0.27–0.97), hospitalization record (aHR = 0.65, 95%CI 0.45–0.93), and higher appointment completion rates (aHR = 0.98, 95% CI 0.98–0.98) were associated with lower risk of death. Overall, and similarly in males, higher age at diagnosis (aHR = 1.03, 95% CI 1.02–1.04) and a TP53 mutation (aHR = 2.24, 95% CI 1.69–2.97) were associated with higher risk of death. Conclusion: Sex differences exist in both AML incidence and overall survival. Treatment and health care factors should be addressed by caregivers and public policies developed to reduce mortality rates and mitigate existing sex differences. [ABSTRACT FROM AUTHOR]

Objective: To determine the frequency of depression or anxiety preceding a diagnosis of pancreatic cancer (PC). Further, to examine the association of PC‐associated depression or anxiety with treatment compliance and survival. Methods: 856 patients with PC from a single institution were identified using International Classification of Diseases (ICD) codes. For each case, two non‐cancer age‐ and sex‐matched controls were included. Dates of depression or anxiety diagnosis identified using ICD codes were compared to the date of PC diagnosis. The medical record was queried to further explore psychiatric symptoms. Multivariable analyses were performed to examine if prediagnosis depression or anxiety was associated with receipt of treatment or survival. Results: A greater proportion of patients with PC experienced depression or anxiety in the year preceding diagnosis than the overall frequency in controls (4.6% vs. 2.6%, p = 0.005) based on ICD codes. Patients with PC exhibited signs of prodromal depression or anxiety based on ICD codes, clinical documentation of psychiatric symptoms, or initiation of new psychiatric medications more often than controls (20.7% vs. 6.7%, p < 0.001). Prediagnosis depression or anxiety was associated with a reduced likelihood of receiving chemotherapy (OR = 0.58, p = 0.04). There was an associated decrease in overall survival among patients with metastatic disease who experienced depression or anxiety before PC diagnosis (HR = 1.32, p = 0.04). Conclusions: The frequency of depression or anxiety among patients with PC was higher than the general population. Prediagnosis psychiatric symptoms were associated with reduced chemotherapy utilization and worse overall survival. Thus, timely identification and treatment of these symptoms may improve outcomes. [ABSTRACT FROM AUTHOR]

The purpose of this study was to examine the effects of different nutrient (carbon, nitrogen, oxygen) concentrations on the microbial activity and community structure in membrane-aerated biofilms (MABs). MABs were grown under well-defined conditions of fluid flow, substrate concentration, and membrane oxygen partial pressure. Biofilms were then removed and thin-sliced using a cryostat/microtome parallel to the membrane. Individual slices were analyzed for changes with depth in biomass density, respiratory activity, and the population densities of ammonia-oxidizing and denitrifying bacteria populations. Oxygen-sensing microelectrodes were used to determine the depth of oxygen penetration into each biofilm. Our results demonstrated that ammonia-oxidizing bacteria grow near the membrane, while denitrifying bacteria grow a substantial distance from the membrane. However, nitrifying and denitrifying bacteria did not grow simultaneously when organic concentrations became too high or ammonia concentrations became too low. In conclusion, membrane-aerated biofilms exhibit substantial stratification with respect to community structure and activity. A fundamental understanding of the factors that control this stratification will help optimize the performance of full-scale membrane-aerated biofilm reactors for wastewater treatment. [ABSTRACT FROM AUTHOR]