Abstract
Objective
This study aims to establish an informative dynamic prediction model of treatment outcomes using follow-up records of tuberculosis (TB) patients, which can timely detect cases when ...the current treatment plan may not be effective.
Materials and Methods
We used 122 267 follow-up records from 17 958 new cases of pulmonary TB in the Republic of Moldova. A dynamic prediction framework integrating landmark modeling and machine learning algorithms was designed to predict patient outcomes during the course of treatment. Sensitivity and positive predictive value (PPV) were calculated to evaluate performance of the model at critical time points. New measures were defined to determine when follow-up laboratory tests should be conducted to obtain most informative results.
Results
The random-forest algorithm performed better than support vector machine and penalized multinomial logistic regression models for predicting TB treatment outcomes. For all 3 outcome classes (ie, cured, not cured, and died after 24 months following treatment initiation), sensitivity and PPV of prediction models improved as more follow-up information was collected. Specifically, sensitivity and PPV increased from 0.55 to 0.84 and from 0.32 to 0.88, respectively, for the not cured class.
Conclusion
The dynamic prediction framework utilizes longitudinal laboratory test results to predict patient outcomes at various landmarks. Sputum culture and smear results are among the important variables for prediction; however, the most recent sputum result is not always the most informative one. This framework can potentially facilitate a more effective treatment monitoring program and provide insights for policymakers toward improved guidelines on follow-up tests.
The effect of breast density on survival outcomes for American women who participate in screening remains unknown. We studied the role of breast density on both breast cancer and other cause of ...mortality in screened women. Data for women with breast cancer, identified from the community-based Carolina Mammography Registry, were linked with the North Carolina cancer registry and NC death tapes for this study. Cause-specific Cox proportional hazards models were developed to analyze the effect of several covariates on breast cancer mortality—namely, age, race (African American/White), cancer stage at diagnosis (in situ, local, regional, and distant), and breast density (BI-RADS
®
1–4). Two stratified Cox models were considered controlling for (1) age and race, and (2) age and cancer stage, respectively, to further study the effect of density. The cumulative incidence function with confidence interval approximation was used to quantify mortality probabilities over time. For this study, 22,597 screened women were identified as having breast cancer. The non-stratified and stratified Cox models showed no significant statistical difference in mortality between dense tissue and fatty tissue, while controlling for other covariate effects (
p
value = 0.1242, 0.0717, and 0.0619 for the non-stratified, race-stratified, and cancer stage–stratified models, respectively). The cumulative mortality probability estimates showed that women with dense breast tissues did not have significantly different breast cancer mortality than women with fatty breast tissue, regardless of age (e.g., 10-year confidence interval of mortality probabilities for whites aged 60–69 white: 0.056–0.090 vs. 0.054–0.083). Aging, African American race, and advanced cancer stage were found to be significant risk factors for breast cancer mortality (hazard ratio >1.0). After controlling for cancer incidence, there was not a significant association between mammographic breast density and mortality, adjusting for the effects of age, race, and cancer stage.
PurposeTeacher pay in Arkansas public schools varies widely from district to district across the state. This pay discrepancy is driven by both the funds available to a district and by how these funds ...are allocated. A standard per student budget is given to districts across the state, but this budget can be supplemented by additional property taxes collected on property within a district. This leaves districts with more highly valued property at an advantage. Districts are free to allocate their budget for teacher pay as they see fit, with constraints on number of students per teacher and minimum teacher salary.Design/methodology/approachUsing public data available through the Arkansas Department of Education, this research investigates what variables affect student performance in Arkansas public schools using feature selection and predictive modelling and determine the cost-effectiveness associated with changing possible decision variables in terms of improving student performance.FindingsIt was found that the most cost-effective ways for districts to increase student performance are to (1) increase average teacher salary and (2) increase average years of teacher experience. This result is validated by education research, as both of these methods have been identified in literature as being effective ways to increase teacher quality and increase student performance. Furthermore, districts should consider increasing student–teacher ratio and applying the resulting savings toward teacher salaries.Originality/valueThis methodology gives a fresh perspective on the most cost-effective use of resources in publicly funded schools.
We consider a variation of an Erlang loss system in which jobs are routed to servers according to the Shortest Idle Server First service discipline. Specifically, we consider a system in which idle ...servers are arranged in a stack; servers are returned to the top of the stack upon service completion; and arriving jobs are assigned to the server currently at the top of the stack. When busy, servers accumulate age and incur an age-dependent operating cost. For such systems, we (i) formulate a continuous-time Markov chain model to characterize the system's transient behavior, and (ii) develop maintenance policies consisting of two possible actions: server group replacement and stack inversion. The stack inversion may be performed at any time prior to group replacement to achieve a more evenly distributed utilization among servers. We develop an optimization model to determine the optimal inversion and replacement times so as to minimize the long-run expected cost rate. Because the model is nonlinear and non-convex, we develop a set of algorithms to solve for the optimal replacement and inversion time. Lastly, we establish a lower bound for the inversion cost threshold below which it is optimal to invert the stack of servers before their replacement.
•A two-stage decision framework is proposed to determine annual mammography decisions based on estimated breast cancer risk.•The derived “online” policy is adaptive to a woman's latest health status, ...which reflects the individual risk of every woman more accurately.•A new model selection method for generalized linear model is proposed that combines tabu search algorithm with the H measure criterion.
The American Cancer Society (ACS) updated their breast cancer screening guidelines in late 2015 and recommends that all women have the choice to start annual mammography screenings beginning at age 40. For women ages 45–54, the ACS explicitly recommends annual mammograms. However, due to the potential harms associated with screening mammography, such as overdiagnosis and unnecessary work-ups, the best strategy to design an appropriate breast cancer mammography screening schedule remains controversial. Instead of recommending a one-size-fits-all screening schedule, this study identifies a personalized mammography screening strategy adaptive to each woman's age-specific breast cancer risk. We present a two-stage decision framework: (1) age-specific breast cancer risk estimation and (2) annual mammography screening decision-making based on estimated risk. The results suggest that the optimal combinations of independent variables used in risk estimation are not the same across age groups. Our optimal decision models outperform the existing mammography screening guidelines in terms of the average loss of life expectancy. While most earlier studies improved the breast cancer screening decisions by offering lifetime screening schedules, our proposed model provides an adaptive screening decision aid by age. Since whether or not a woman should receive a mammogram is determined based on her breast cancer risk at her current age, our “on-line” screening policy adapts to a woman's latest health status, which reflects the current individual risk of each woman more accurately.
Background
Human papillomavirus (HPV) is the most common sexually transmitted infection in the United States. HPV can cause genital warts and multiple types of cancers in females. HPV vaccination is ...recommended to youth age 11 or 12 years before sexual initiation to prevent onset of HPV-related diseases. For females who have not been vaccinated previously, catch-up vaccines are recommended through age 26. The extent to which catch-up vaccines are beneficial in terms of disease prevention and cost-effectiveness is questionable given that some women may have been exposed to HPV before receiving the catch-up vaccination. This study aims to examine whether the cutoff age of catch-up vaccination should be determined based on an individual woman’s risk characteristic instead of a one-size-fits-all age 26.
Methods
We developed a microsimulation model to evaluate multiple clinical outcomes of HPV vaccination for different women based on a number of personal attributes. We modeled the impact of HPV vaccination at different ages on every woman and tracked her course of life to estimate the clinical outcomes that resulted from receiving vaccines. As the simulation model is risk stratified, we used extreme gradient boosting to build an HPV risk model estimating every woman’s dynamic HPV risk over time for the lifetime simulation model.
Results
Our study shows that catch-up vaccines still benefit all women after age 26 from the perspective of clinical outcomes. Women facing high risk of HPV infection are expected to gain more health benefits compared with women with low HPV risk.
Conclusions
From a cancer prevention perspective, this study suggests that the catch-up vaccine after age 26 should be deliberately considered.
Machine vision-based food quality evaluation systems have achieved great attention in industry and academia in recent years because of their high-throughput and non-invasive properties. In practice, ...the environmental illumination conditions variations would cause visual evaluation bias for both human and machine perceptions. Compared to other existing studies which take sample images and human visual grading under fixed illumination conditions, this study first investigated the environmental illumination effects on the performance of visual-based food grading for both humans and machines. Taking lettuce samples as an example, an image dataset that considered the environmental illumination variations was first established. This dataset encompasses human visual grading scores obtained from sensory panels as well. In contrast to current studies that utilize a single grader or L*a*b* color features to assess sample freshness, our study reveals a substantial impact of environmental illumination on both human perception (p<0.0001) and L*a*b* color features (p<0.0001). In order to enhance the performance of machine vision-based freshness prediction, multitask learning protocols were incorporated into the proposed new network architectures. This allowed the simultaneous prediction of both sample freshness and illumination conditions. In comparison to commonly used generic convolutional neural network models and vision transformer models, the newly proposed model exhibited superior freshness prediction performance. It minimized the prediction error by 20.36%, outperforming the generic ResNet model. This research represents the first quantitative study addressing human and machine perceptions under varied illumination conditions for food quality evaluation. The findings are anticipated to play a pivotal role in expediting the integration of machine vision applications into food engineering practices.
Display omitted
•Built the first food quality evaluation dataset for illumination robustness studies.•Firstly evaluate human and machine perception robustness to different illumination.•Firstly benchmark off-the-shelf vision models for food freshness predictions.•Introduce auxiliary illumination predictions to improve the model performances.
This study quantifies breast cancer mortality in the presence of competing risks for complex patients. Breast cancer behaves differently in different patient populations, which can have significant ...implications for patient survival; hence these differences must be considered when making screening and treatment decisions. Mortality estimation for breast cancer patients has been a significant research question. Accurate estimation is critical for clinical decision making, including recommendations. In this study, a competing risks framework is built to analyze the effect of patient risk factors and cancer characteristics on breast cancer and other cause mortality. To estimate mortality probabilities from breast cancer and other causes as a function of not only the patient’s age or race but also biomarkers for estrogen and progesterone receptor status, a nonparametric cumulative incidence function is formulated using data from the community-based Carolina Mammography Registry. Based on the log(−log) transformation, confidence intervals are constructed for mortality estimates over time. To compare mortality probabilities in two independent risk groups at a given time, a method with improved power is formulated using the log(−log) transformation.
•We incorporate women's nonstationary adherence behavior into breast cancer screening decision modeling.•We develop a partially observable Markov chain model to incorporate imperfectness of screening ...tests.•Both screening mammography and breast self examination are considered as methods of breast cancer detection.•Interval cancer is incorporated in the model.•Our results can facilitate healthcare providers to tailor screening recommendations for patients.
Mammography is known to be the most effective way of breast cancer detection. The efficacy of mammography screening guidelines is highly associated with women's compliance with these recommendations. Currently, none of the existing policies takes women's behavior into consideration; instead, perfect adherence is assumed. However, an earlier longitudinal data analysis has revealed that women's compliance with mammography guidelines remained low in recent years. In this study, we develop a randomized discrete-time finite-horizon partially observable Markov chain model to evaluate a wide range of screening mammography policies, incorporating heterogeneity in women's adherence behaviors. Considering potential harms of mammography tests (e.g., risk of developing radiation-induced breast cancer, false negatives, false positives, etc.), policies with varying starting age, ending age and frequency of screening mammograms at different ages are compared in terms of total quality adjusted life years (QALYs) and lifetime breast cancer mortality risk. Our results show that women with perfect adherence do not always experience higher QALYs. In fact, for some policies, including the American Cancer Society (ACS) policy, the general population with various adherence levels has higher QALYs than women with perfect adherence. However, in terms of lifetime breast cancer mortality risk, higher/perfect adherence always results in lower risk of dying from breast cancer. This implies that the benefits of mammography in decreasing death from breast cancer outweigh the increased risk of developing radiation-induced breast cancer from mammographic screening. This study can facilitate healthcare providers to tailor screening mammography recommendations based on their patients' estimated adherence levels.
Background
Lung volume reduction surgery (LVRS) and medical therapy are 2 available treatment options in dealing with severe emphysema, which is a chronic lung disease. However, or there are ...currently limited guidelines on the timing of LVRS for patients with different characteristics.
Objective
The objective of this study is to assess the timing of receiving LVRS in terms of patient outcomes, taking into consideration a patient’s characteristics.
Methods
A finite-horizon Markov decision process model for patients with severe emphysema was developed to determine the short-term (5 y) and long-term timing of emphysema treatment. Maximizing the expected life expectancy, expected quality-adjusted life-years, and total expected cost of each treatment option were applied as the objective functions of the model. To estimate parameters in the model, the data provided by the National Emphysema Treatment Trial were used.
Results
The results indicate that the treatment timing strategy for patients with upper-lobe predominant emphysema is to receive LVRS regardless of their specific characteristics. However, for patients with non–upper-lobe–predominant emphysema, the optimal strategy depends on the age, maximum workload level, and forced expiratory volume in 1 second level.
Conclusion
This study demonstrates the utilization of clinical trial data to gain insights into the timing of surgical treatment for patients with emphysema, considering patient age, observable health condition, and location of emphysema.
Highlights
Both short-term and long-term Markov decision process models were developed to assess the timing of receiving lung volume reduction surgery in patients with severe emphysema.
How clinical trial data can be used to estimate the parameters and obtain short-term results from the Markov decision process model is demonstrated.
The results provide insights into the timing of receiving lung volume reduction surgery as a function of a patient’s characteristics, including age, emphysema location, maximum workload, and forced expiratory volume in 1 second level.