Boosting the diagnostic power of amyloid-β PET using a data-driven spatially informed classifier for decision support
Ontology highlight
ABSTRACT: Background
Amyloid-β (Aβ) PET has emerged as clinically useful for more accurate diagnosis of patients with cognitive decline. Aβ deposition is a necessary cause or response to the cellular pathology of Alzheimer’s disease (AD). Usual clinical and research interpretation of amyloid PET does not fully utilise all information regarding the spatial distribution of signal. We present a data-driven, spatially informed classifier to boost the diagnostic power of amyloid PET in AD. Methods
Voxel-wise k-means clustering of amyloid-positive voxels was performed; clusters were mapped to brain anatomy and tested for their associations by diagnostic category and disease severity with 758 amyloid PET scans from volunteers in the AD continuum from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). A machine learning approach based on this spatially constrained model using an optimised quadratic support vector machine was developed for automatic classification of scans for AD vs non-AD pathology. Results
This classifier boosted the accuracy of classification of AD scans to 81% using the amyloid PET alone with an area under the curve (AUC) of 0.91 compared to other spatial methods. This increased sensitivity to detect AD by 15% and the AUC by 9% compared to the use of a composite region of interest SUVr. Conclusions
The diagnostic classification accuracy of amyloid PET was improved using an automated data-driven spatial classifier. Our classifier highlights the importance of considering the spatial variation in Aβ PET signal for optimal interpretation of scans. The algorithm now is available to be evaluated prospectively as a tool for automated clinical decision support in research settings. Supplementary Information
The online version contains supplementary material available at 10.1186/s13195-021-00910-8.
SUBMITTER: Venkataraman A
PROVIDER: S-EPMC8582159 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundPublic health professionals are increasingly expected to engage in evidence-informed decision making to inform practice and policy decisions. Evidence-informed decision making involves the use of research evidence along with expertise, existing public health resources, knowledge about community health issues, the local context and community, and the political climate. The National Collaborating Centre for Methods and Tools has identified a seven step process for evidence-informed decision making. Tools have been developed to support public health professionals as they work through each of these steps. This paper provides an overview of tools used in three Canadian public health departments involved in a study to develop capacity for evidence-informed decision making.MethodsAs part of a knowledge translation and exchange intervention, a Knowledge Broker worked with public health professionals to identify and apply tools for use with each of the steps of evidence-informed decision making. The Knowledge Broker maintained a reflective journal and interviews were conducted with a purposive sample of decision makers and public health professionals. This paper presents qualitative analysis of the perceived usefulness and usability of the tools.ResultsTools were used in the health departments to assist in: question identification and clarification; searching for the best available research evidence; assessing the research evidence for quality through critical appraisal; deciphering the 'actionable message(s)' from the research evidence; tailoring messages to the local context to ensure their relevance and suitability; deciding whether and planning how to implement research evidence in the local context; and evaluating the effectiveness of implementation efforts. Decision makers provided descriptions of how the tools were used within the health departments and made suggestions for improvement. Overall, the tools were perceived as valuable for advancing and sustaining evidence-informed decision making.ConclusionTools are available to support the process of evidence-informed decision making among public health professionals. The usability and usefulness of these tools for advancing and sustaining evidence-informed decision making are discussed, including recommendations for the tools' application in other public health settings beyond this study. Knowledge and awareness of these tools may assist other health professionals in their efforts to implement evidence-informed practice.
| S-EPMC4223550 | biostudies-literature
Project description:Rapidly improving understanding of molecular oncology, emerging novel therapeutics, and increasingly available and affordable next-generation sequencing have created an opportunity for delivering genomically informed personalized cancer therapy. However, to implement genomically informed therapy requires that a clinician interpret the patient's molecular profile, including molecular characterization of the tumor and the patient's germline DNA. In this Commentary, we review existing data and tools for precision oncology and present a framework for reviewing the available biomedical literature on therapeutic implications of genomic alterations. Genomic alterations, including mutations, insertions/deletions, fusions, and copy number changes, need to be curated in terms of the likelihood that they alter the function of a "cancer gene" at the level of a specific variant in order to discriminate so-called "drivers" from "passengers." Alterations that are targetable either directly or indirectly with approved or investigational therapies are potentially "actionable." At this time, evidence linking predictive biomarkers to therapies is strong for only a few genomic markers in the context of specific cancer types. For these genomic alterations in other diseases and for other genomic alterations, the clinical data are either absent or insufficient to support routine clinical implementation of biomarker-based therapy. However, there is great interest in optimally matching patients to early-phase clinical trials. Thus, we need accessible, comprehensive, and frequently updated knowledge bases that describe genomic changes and their clinical implications, as well as continued education of clinicians and patients.
| S-EPMC4651038 | biostudies-literature
Project description:We present new methodology based on Multi-Objective Markov Decision Processes for developing sequential decision support systems from data. Our approach uses sequential decision-making data to provide support that is useful to many different decision-makers, each with different, potentially time-varying preference. To accomplish this, we develop an extension of fitted-Q iteration for multiple objectives that computes policies for all scalarization functions, i.e. preference functions, simultaneously from continuous-state, finite-horizon data. We identify and address several conceptual and computational challenges along the way, and we introduce a new solution concept that is appropriate when different actions have similar expected outcomes. Finally, we demonstrate an application of our method using data from the Clinical Antipsychotic Trials of Intervention Effectiveness and show that our approach offers decision-makers increased choice by a larger class of optimal policies.
| S-EPMC5179144 | biostudies-literature
Project description:An essential component of therapeutic drug/biomarker monitoring (TDM) is to combine patient data with prior knowledge for model-based predictions of therapy outcomes. Current Bayesian forecasting tools typically rely only on the most probable model parameters (maximum a posteriori (MAP) estimate). This MAP-based approach, however, does neither necessarily predict the most probable outcome nor does it quantify the risks of treatment inefficacy or toxicity. Bayesian data assimilation (DA) methods overcome these limitations by providing a comprehensive uncertainty quantification. We compare DA methods with MAP-based approaches and show how probabilistic statements about key markers related to chemotherapy-induced neutropenia can be leveraged for more informative decision support in individualized chemotherapy. Sequential Bayesian DA proved to be most computationally efficient for handling interoccasion variability and integrating TDM data. For new digital monitoring devices enabling more frequent data collection, these features will be of critical importance to improve patient care decisions in various therapeutic areas.
| S-EPMC7080550 | biostudies-literature
Project description:ObjectivesPhysician-facing decision support tools facilitate shared decision-making (SDM) during informed consent, but it is unclear whether they are comprehensive in the domains they measure. In this scoping review, we aimed to (1) identify the physician-facing tools used during SDM; (2) assess the patient-centered domains measured by these tools; (3) determine whether tools are available for older adults and for use in various settings (elective vs. emergent); and (4) characterize domains future tools should measure.MethodsUsing the Preferred Reporting Items for Systematic Reviews and Meta-analyses extension for Scoping Reviews, Embase, Medline, and Web of Science were queried for articles published between January 2000 and September 2022. Articles meeting inclusion criteria underwent title and abstract review. Eligible studies underwent data abstraction by two reviewers.ResultsOf 4365 articles identified, 160 were eligible. Tools to aid in surgical SDM focus on elective procedures (79%) and the outpatient setting (71%). Few tools are designed for older adults (5%) or for nonelective procedures (9%). Risk calculators were most common, followed by risk indices, prognostic nomograms, and communication tools. Of the domains measured, prognosis was more commonly measured (85%), followed by alternatives (28%), patient goals (36%), and expectations (46%). Most tools represented only one domain (prognosis, 33.1%) and only 6.7% represented all four domains.Conclusions and implicationsTools to aid in the surgical SDM process measure short-term prognosis more often than patient-centered domains such as long-term prognosis, patient goals, and expectations. Further research should focus on communication tools, the needs of older patients, and use in diverse settings.
| S-EPMC10431347 | biostudies-literature
Project description:BackgroundMinistries of health in low- and middle-income countries often lack timely quality data for data-driven decision making in healthcare networks. We describe the design and implementation of decision-support electronic tools by the Ministry of Health of the State of Chiapas, in Mexico, as part of Salud Mesoamerica Initiative.MethodsThree electronic decision-support tools were designed through an iterative process focused on streamlined implementation: 1) to collect and report health facility data at health facilities; 2) to compile and analyze data at health district and central level; and, 3) to support stratified sampling of health facilities. Data was collected for five composite indicators measuring availability of equipment, medicines, and supplies for maternal and child health. Quality Assurance Teams collected data, evaluated results and supported quality improvement. Data was also analyzed at the central level and health districts for decision-making.ResultsData from 300 health facilities in four health districts was collected and analyzed (November 2014-June 2015). The first wave revealed gaps on availability of equipment and supplies in more than half of health facilities. Electronic tools provided the ministry of health officers new ways to visualize data, identify patterns and make hypothesis on root-causes. Between the first and second measurement, the number of missing items decreased, and actions performed by quality improvement teams became more proactive. In the final measurement, 89.7-100% of all health facilities achieved all the required items for each indicator.ConclusionsOur experience could help guide others seeking to implement electronic decision-support tools in low- and middle-income countries. Electronic decision-support tools supported data-driven decision-making by identifying gaps on heatmaps and graphs at the health facility, subdistrict, district or state level. Through a rapid improvement process, the Ministry of Health met targets of externally verified indicators. Using available information technology resources facilitated prompt implementation and adoption of technology.
| S-EPMC7301503 | biostudies-literature
Project description:BackgroundWith opioid misuse, opioid use disorder (OUD), and opioid overdose deaths persisting at epidemic levels in the U.S., the largest implementation study in addiction research-the HEALing Communities Study (HCS)-is evaluating the impact of the Communities That Heal (CTH) intervention on reducing opioid overdose deaths in 67 disproportionately affected communities from four states (i.e., "sites"). Community-tailored dashboards are central to the CTH intervention's mandate to implement a community-engaged and data-driven process. These dashboards support a participating community's decision-making for selection and monitoring of evidence-based practices to reduce opioid overdose deaths.Methods/designA community-tailored dashboard is a web-based set of interactive data visualizations of community-specific metrics. Metrics include opioid overdose deaths and other OUD-related measures, as well as drivers of change of these outcomes in a community. Each community-tailored dashboard is a product of a co-creation process between HCS researchers and stakeholders from each community. The four research sites used a varied set of technical approaches and solutions to support the scientific design and CTH intervention implementation. Ongoing evaluation of the dashboards involves quantitative and qualitative data on key aspects posited to shape dashboard use combined with website analytics.DiscussionThe HCS presents an opportunity to advance how community-tailored dashboards can foster community-driven solutions to address the opioid epidemic. Lessons learned can be applied to inform interventions for public health concerns and issues that have disproportionate impact across communities and populations (e.g., racial/ethnic and sexual/gender minorities and other marginalized individuals).Trial registrationClinicalTrials.gov (NCT04111939).
| S-EPMC7528750 | biostudies-literature
Project description:Two major treatment strategies employed in non-small cell lung cancer, NSCLC, are tyrosine kinase inhibitors, TKIs, and immune checkpoint inhibitors, ICIs. The choice of strategy is based on heterogeneous biomarkers that can dynamically change during therapy. Thus, there is a compelling need to identify comprehensive biomarkers that can be used longitudinally to help guide therapy choice. Herein, we report a 18F-FDG-PET/CT-based deep learning model, which demonstrates high accuracy in EGFR mutation status prediction across patient cohorts from different institutions. A deep learning score (EGFR-DLS) was significantly and positively associated with longer progression free survival (PFS) in patients treated with EGFR-TKIs, while EGFR-DLS is significantly and negatively associated with higher durable clinical benefit, reduced hyperprogression, and longer PFS among patients treated with ICIs. Thus, the EGFR-DLS provides a non-invasive method for precise quantification of EGFR mutation status in NSCLC patients, which is promising to identify NSCLC patients sensitive to EGFR-TKI or ICI-treatments.
| S-EPMC7567795 | biostudies-literature
Project description:Advanced high-throughput screening (HTS) technologies generate great amounts of bioactivity data, and this data needs to be analyzed and interpreted with attention to understand how these small molecules affect biological systems. As such, there is an increasing demand to develop and adapt cheminformatics algorithms and tools in order to predict molecular and pharmacological properties on the basis of these large data sets. In this manuscript, we report a novel machine-learning-based ligand classification algorithm, named Ligand Classifier of Adaptively Boosting Ensemble Decision Stumps (LiCABEDS), for data-mining and modeling of large chemical data sets to predict pharmacological properties in an efficient and accurate manner. The performance of LiCABEDS was evaluated through predicting GPCR ligand functionality (agonist or antagonist) using four different molecular fingerprints, including Maccs, FP2, Unity, and Molprint 2D fingerprints. Our studies showed that LiCABEDS outperformed two other popular techniques, classification tree and Naive Bayes classifier, on all four types of molecular fingerprints. Parameters in LiCABEDS, including the number of boosting iterations, initialization condition, and a "reject option" boundary, were thoroughly explored and discussed to demonstrate the capability of handling imbalanced data sets, as well as its robustness and flexibility. In addition, the detailed mathematical concepts and theory are also given to address the principle behind statistical prediction models. The LiCABEDS algorithm has been implemented into a user-friendly software package that is accessible online at http://www.cbligand.org/LiCABEDS/ .
| S-EPMC3065508 | biostudies-literature
Project description:With the ever-increasing societal dependence on electricity, one of the critical tasks in power supply is maintaining the power line infrastructure. In the process of making informed, cost-effective, and timely decisions, maintenance engineers must rely on human-created, heterogeneous, structured, and also largely unstructured information. The maturing research on vision-based power line inspection driven by advancements in deep learning offers first possibilities to move towards more holistic, automated, and safe decision-making. However, (current) research focuses solely on the extraction of information rather than its implementation in decision-making processes. The paper addresses this shortcoming by designing, instantiating, and evaluating a holistic deep-learning-enabled image-based decision support system artifact for power line maintenance at a German distribution system operator in southern Germany. Following the design science research paradigm, two main components of the artifact are designed: A deep-learning-based model component responsible for automatic fault detection of power line parts as well as a user-oriented interface responsible for presenting the captured information in a way that enables more informed decisions. As a basis for both components, preliminary design requirements are derived from literature and the application field. Drawing on justificatory knowledge from deep learning as well as decision support systems, tentative design principles are derived. Based on these design principles, a prototype of the artifact is implemented that allows for rigorous evaluation of the design knowledge in multiple evaluation episodes, covering different angles. Through a technical experiment the technical novelty of the artifact’s capability to capture selected faults (regarding insulators and safety pins) in unmanned aerial vehicle (UAV)-captured image data (model component) is validated. Subsequent interviews, surveys, and workshops in a natural environment confirm the usefulness of the model as well as the user interface component. The evaluation provides evidence that (1) the image processing approach manages to address the gap of power line component inspection and (2) that the proposed holistic design knowledge for image-based decision support systems enables more informed decision-making. The paper therefore contributes to research and practice in three ways. First, the technical feasibility to detect certain maintenance-intensive parts of power lines with the help of unique UAV image data is shown. Second, the distribution system operators’ specific problem is solved by supporting decisions in maintenance with the proposed image-based decision support system. Third, precise design knowledge for image-based decision support systems is formulated that can inform future system designs of a similar nature. Supplementary Information The online version contains supplementary material available at 10.1007/s12599-022-00745-z.
| S-EPMC8973684 | biostudies-literature