We present a data-driven approach to characterizing nonidentifiability of a model’s parameters and illustrate it through dynamic as well as steady kinetic models. By employing Diffusion Maps and their extensions, we ...
详细信息
The accelerated adoption of digital pathology and advances in deep learning have enabled the development of powerful models for various pathology tasks across a diverse array of diseases and patient cohorts1–13. Howe...
详细信息
Symbolic regression searches for analytic expressions that accurately describe studied phenomena. The main promise of this approach is that it may return an interpretable model that can be insightful to users, while m...
详细信息
Symbolic regression searches for analytic expressions that accurately describe studied phenomena. The main promise of this approach is that it may return an interpretable model that can be insightful to users, while maintaining high accuracy. The current standard for benchmarking these algorithms is SRBench, which evaluates methods on hundreds of datasets that are a mix of real-world and simulated processes spanning multiple domains. At present, the ability of SRBench to evaluate interpretability is limited to measuring the size of expressions on real-world data, and the exactness of model forms on synthetic data. In practice, model size is only one of many factors used by subject experts to determine how interpretable a model truly is. Furthermore, SRBench does not characterize algorithm performance on specific, challenging sub-tasks of regression such as feature selection and evasion of local minima. In this work, we propose and evaluate an approach to benchmarking SR algorithms that addresses these limitations of SRBench by 1) incorporating expert evaluations of interpretability on a domain-specific task, and 2) evaluating algorithms over distinct properties of data science tasks. We evaluate 12 modern symbolic regression algorithms on these benchmarks and present an in-depth analysis of the results, discuss current challenges of symbolic regression algorithms and highlight possible improvements for the benchmark itself. Authors
AbstractBackgroundThe prevalence of missing data in the National Cancer Database (NCDB) has marked implications on clinical care and research. The objective of this study was to enhance the NCDB by decreasing rates of...
详细信息
AbstractBackgroundThe prevalence of missing data in the National Cancer Database (NCDB) has marked implications on clinical care and research. The objective of this study was to enhance the NCDB by decreasing rates of missingness and adding new variables using automated statistical methodology. MethodsOne health system’s NCDB data from 2011–2021 was linked to electronic health record (EHR). Variables with frequent missingness and new clinically significant variables not yet included in the NCDB including patient Eastern Cooperative Oncology Group (ECOG) score, specific chemotherapy regimen, American Society of Anesthesiologists Physical Status Classification (ASA class), and discrete surgical procedure were identified in structured and unstructured EHR data. After automated incorporation of structured data from EHR, a natural language processing tool incorporating rule-based algorithms was designed to further extract variables from unstructured notes. Rates of missingness were compared between the original NCDB and the enhanced dataset, and example multivariable models were run to assess for altered model performance with reduced missingness and the addition of new clinically significant variables (chemotherapy regimen). ResultsA total of 6050 patients with NCDB records were linked to their EHR data. Prior to enhancement, rates of missingness for key variables ranged from 2.0% to 5.3%. Following dataset enhancement, missingness was significantly reduced, with relative missingness being reduced between 31.9% to 68.0%. Of the new variables added, 1367 (22.6%) of 6050 patients gained ECOG score, and 1099 (57.8%) of 1901 who received chemotherapy gained their chemotherapy regimen. Of 2989 who underwent surgery, 979 (32.8%) gained their procedure name and 621 (20.8%) gained ASA class. Comparison of the multivariable models demonstrated significant differences between the original NCDB and the enhanced dataset. Specifically, when replacing the binary predictor for chemoth
Proteins fold to a specific functional conformation with a densely packed core that controls their stability. Despite their importance, we lack a quantitative explanation for why all protein cores, regardless of their...
Proteins fold to a specific functional conformation with a densely packed core that controls their stability. Despite their importance, we lack a quantitative explanation for why all protein cores, regardless of their overall fold, possess the same average packing fraction 〈ϕ〉≈0.55. However, important developments in the physics of jamming in particulate systems can shed light on the packing of protein cores. Here, we extend the framework of jamming to describe core packing in collapsed polymers, as well as in all-atom models of folded proteins. First, we show in a spherical bead-spring polymer model (with and without bond-angle constraints) that as the hydrophobic interactions increase relative to thermal fluctuations, a jamming-like transition occurs when the core packing fraction exceeds ϕc with the same power-law scaling behavior for the potential energy Vr, excess contact number ΔN, and characteristic frequency of the vibrational density of states ω* versus Δϕ=ϕ−ϕc as that for jammed particulate systems. Then, we develop an all-atom model for proteins and find that, above ϕc∼0.55, protein cores undergo a jamming-like transition, but with anomalous power-law scaling for Vr, ΔN, and ω* versus Δϕ. The all-atom protein model remains close to the native protein structure during jamming and accurately refolds from partially unfolded states.
Background: Within the framework of a randomized controlled trial investigating the impact of a digital, psychosocial photo activity intervention for residents living with dementia in nursing homes and their informal ...
详细信息
Background: Within the framework of a randomized controlled trial investigating the impact of a digital, psychosocial photo activity intervention for residents living with dementia in nursing homes and their informal and formal carers, a process evaluation was conducted to determine factors that affected the implementation of the intervention and potentially influenced the intervention outcomes. Objective: By tracing facilitators and barriers to implementation, the study also aimed to inform future implementation of the photo activity intervention. Methods: Following Medical Research Council guidance, mixed methods were used to investigate context, implementation, and mechanism-of-impact factors during the photo activity intervention via the Fotoscope web application versus a general conversation activity (control). Google Analytics was set up to gain insight into how the Fotoscope web application was used in practice. For quantitative data, descriptive statistics were calculated and differences between groups tested. For qualitative data, thematic analysis was performed. Results: In total, 163 semistructured interviews were conducted with residents (photo activity group: n=29, 17.8%;control: n=29, 17.8%), formal carers (photo activity group: n=23, 14.1%;control: n=27, 16.6%), and informal carers (photo activity group: n=28, 17.2%;control: n=27, 16.6%). Regarding contextual factors, a minority of formal carers in both groups (photo activity group: 4/18, 22%;control: 9/24, 38%) mentioned time and workload as barriers to implementing the intervention. Regarding implementation, 86% (25/29) of the residents in the intervention group felt that the digital photo activity worked well on a tablet. Informal carers from both groups wanted more intervention updates from formal carers. The majority of formal carers from both groups were satisfied with how the training and activities were implemented. Regarding the mechanisms of impact, residents in the photo activity group (27/
Smart Fish Feeder is an automatic fish feeder that can be controlled using android smartphone. With this tool, fish owners will easily adjust the feeding schedule according to the recommended feed dose, and provide aq...
详细信息
ISBN:
(数字)9781728120942
ISBN:
(纸本)9781728120959
Smart Fish Feeder is an automatic fish feeder that can be controlled using android smartphone. With this tool, fish owners will easily adjust the feeding schedule according to the recommended feed dose, and provide aquarium cleaning scheduling. In designing a fish feeder, there are some criteria such as temperature and time interval of feeding. The data collection has been done by interviewing ornamental fish traders. The data also has been obtained from literature studies that support problem solving theory. The Laravel framework is used to interpret the system while Firebase as the Database Management System. The Android is as a front end that interacts directly with the user. Automatic fish feeding systems are implemented using arduino micro-controller and prototype feeding devices. This research uses Fuzzy Logic Controller method. With the creation of the prototype smart fish feeder the device functions well in terms of both controller and push data. The results of the calculation of the duration of fish feed using the Fuzzy Sugeno Algorithm have been successfully applied to the smart fish feeder.
Symbolic Regression (SR) is a powerful technique for discovering interpretable mathematical expressions. However, benchmarking SR methods remains challenging due to the diversity of algorithms, datasets, and evaluatio...
详细信息
SG technology is becoming a reality, with several deployments appearing around the world. Despite bringing improvements, 5G technology may not adequately support many envisaged applications that will demand large amou...
详细信息
ISBN:
(数字)9781728174075
ISBN:
(纸本)9781728174082
SG technology is becoming a reality, with several deployments appearing around the world. Despite bringing improvements, 5G technology may not adequately support many envisaged applications that will demand large amounts of network resources. Thus, studies on the next generation, the 6G, are highly desirable. Observing this landscape, in this paper, we present a seminal study about the potential usage of 6G to support such highly demanding applications. Towards this study, we model a 6G system at the simulation level and conduct a case study exploring an application of high-definition video monitoring, applying the concepts of drone-swarm based surveillance. This scenario demands the transmission of large amounts of video data through the network. The acquired results present evidence that 6G can address these high network traffic demands.
The Common European Framework of Reference (CEFR) guidelines describe language proficiency of learners on a scale of 6 levels. While the description of CEFR guidelines is generic across languages, the development of a...
详细信息
暂无评论