Modeling Pipeline Research Articles

Enhancing Mental Fatigue Detection through Physiological Signals and Machine Learning Using Contextual Insights and Efficient Modelling

This research presents a machine learning modeling process for detecting mental fatigue using three physiological signals: electrodermal activity, electrocardiogram, and respiration. It follows the conventional machine learning modeling pipeline, while emphasizing the significant contribution of the feature selection process, resulting in, not only a high-performance model, but also a relevant one. The employed feature selection process considers both statistical and contextual aspects of feature relevance. Statistical relevance was assessed through variance and correlation analyses between independent features and the dependent variable (fatigue state). A contextual analysis was based on insights derived from the experimental design and feature characteristics. Additionally, feature sequencing and set conversion techniques were employed to incorporate the temporal aspects of physiological signals into the training of machine learning models based on random forest, decision tree, support vector machine, k-nearest neighbors, and gradient boosting. An evaluation was conducted using a dataset acquired from a wearable electronic system (in third-party research) with physiological data from three subjects undergoing a series of tests and fatigue stages. A total of 18 tests were performed by the 3 subjects in 3 mental fatigue states. Fatigue assessment was based on subjective measures and reaction time tests, and fatigue induction was performed through mental arithmetic operations. The results showed the highest performance when using random forest, achieving an average accuracy and F1-score of 96% in classifying three levels of mental fatigue.

Robust cosmological inference from non-linear scales with k-th nearest neighbour statistics

ABSTRACT We present the methodology for deriving accurate and reliable cosmological constraints from non-linear scales ($\lt 50\, h^{-1}$ Mpc) with k-th nearest neighbour (kNN) statistics. We detail our methods for choosing robust minimum scale cuts and validating galaxy–halo connection models. Using cross-validation, we identify the galaxy–halo model that ensures both good fits and unbiased predictions across diverse summary statistics. We demonstrate that we can model kNNs effectively down to transverse scales of $r_{\rm p}\sim 3\, h^{-1}$ Mpc and achieve precise and unbiased constraints on the matter density and clustering amplitude, leading to a 2 per cent constraint on σ8. Our simulation-based model pipeline is resilient to varied model systematics, spanning simulation codes, halo finding, and cosmology priors. We demonstrate the effectiveness of this approach through an application to the Beyond-2p mock challenge. We propose further explorations to test more complex galaxy–halo connection models and tackle potential observational systematics.

FitMultiCell: simulating and parameterizing computational models of multi-scale and multi-cellular processes.

Biological tissues are dynamic and highly organized. Multi-scale models are helpful tools to analyse and understand the processes determining tissue dynamics. These models usually depend on parameters that need to be inferred from experimental data to achieve a quantitative understanding, to predict the response to perturbations, and to evaluate competing hypotheses. However, even advanced inference approaches such as approximate Bayesian computation (ABC) are difficult to apply due to the computational complexity of the simulation of multi-scale models. Thus, there is a need for a scalable pipeline for modeling, simulating, and parameterizing multi-scale models of multi-cellular processes. Here, we present FitMultiCell, a computationally efficient and user-friendly open-source pipeline that can handle the full workflow of modeling, simulating, and parameterizing for multi-scale models of multi-cellular processes. The pipeline is modular and integrates the modeling and simulation tool Morpheus and the statistical inference tool pyABC. The easy integration of high-performance infrastructure allows to scale to computationally expensive problems. The introduction of a novel standard for the formulation of parameter inference problems for multi-scale models additionally ensures reproducibility and reusability. By applying the pipeline to multiple biological problems, we demonstrate its broad applicability, which will benefit in particular image-based systems biology. FitMultiCell is available open-source at https://gitlab.com/fitmulticell/fit.

Framework for federated causal inference based on real-world observational data sources (BY-COVID)

Abstract Background Causal inference techniques help researchers and policy-makers to evaluate public health interventions. Approaching causal inference by re-using routinely collected observational data across different regions is challenging and guidance is currently lacking. With the aim of filling this gap and allowing a rapid response in the case of a next pandemic, a methodological framework to estimate causal effects in a federated research infrastructure is showcased in the baseline use case of the European BY-COVID project. Methods A framework, implementing existing methodologies and expertise, is proposed, enabling federated research based on routinely collected sensitive individual-level data. The framework includes step-by-step guidelines, from defining a research question, to establishing a causal model, identifying and specifying data requirements in a data model, and developing an interoperable and reproducible analytical pipeline for distributed deployment. Using open-source software, a complete workflow implementing federated causal inference was prototyped to assess the real-world effectiveness of SARS-CoV-2 primary vaccination in preventing infection in populations spanning different countries, integrating a data quality assessment, imputation of missing values, matching of exposed to unexposed based on identified confounders and a survival analysis within the matched population. Results The proposed methodological framework was successfully demonstrated within the BY-COVID use case, illustrating its feasibility and value. Different Findable, Accessible, Interoperable and Reusable (FAIR) research objects were produced, such as a study protocol, a data management plan, data model and interoperable analytical pipeline. Conclusions The framework provides a systematic approach to address policy-relevant causal research questions in a privacy-preserving way. The methodology and derived research objects can be re-used and contribute to pandemic preparedness. Key messages • Estimating causal effects using observational data in a privacy-preserving way was demonstrated to be achievable and can help policy-makers to evaluate public health interventions. • The methodology, constituted of Findable, Accessible, Interoperable and Reusable (FAIR) research objects, is generalizable to other research questions and contributes to pandemic preparedness.

A data-driven computational model for engineered cardiac microtissues

Engineered heart tissues (EHTs) present a potential solution to some of the current challenges in the treatment of heart disease; however, the development of mature, adult-like cardiac tissues remains elusive. Mechanical stimuli have been observed to improve whole-tissue function and cardiomyocyte (CM) maturation, although our ability to fully utilize these mechanisms is hampered, in part, by our incomplete understanding of the mechanobiology of EHTs. In this work, we leverage experimental data, produced by a mechanically tunable experimental setup, to introduce a tissue-specific computational modeling pipeline of EHTs. Our new modeling pipeline generates simulated, image-based EHTs, capturing ECM and myofibrillar structure as well as functional parameters estimated directly from experimental data. This approach enables the unique estimation of EHT function by data-based estimation of CM active stresses. We use this experimental and modeling pipeline to study different mechanical environments, where we contrast the force output of the tissue with the computed active stress of CMs. We show that the significant differences in measured experimental forces can largely be explained by the levels of myofibril formation achieved by the CMs in the distinct mechanical environments, with active stress showing more muted variations across conditions. The presented model also enables us to dissect the relative contributions of myofibrils and extracellular matrix to tissue force output, a task difficult to address experimentally. These results highlight the importance of tissue-specific modeling to augment EHT experiments, providing deeper insights into the mechanobiology driving EHT function. Statement of significanceEngineered heart tissues (EHTs) have the potential to revolutionize the way heart disease is treated. However, developing mature cardiomyocytes (CM) in these tissues remains a challenge due, in part, to our incomplete understanding of the fundamental biomechanical mechanisms that drive EHT development. This work integrates the experimental data of an EHT platform developed to study the influence of mechanics in CM maturation with computational biomechanical models. This approach is used to augment conclusions obtained in-vitro - by measuring quantities such as cell stress and strain - and to dissect the relevance of each component in the whole tissue performance. Our results show how a combination of specialized in-silico and in-vitro approaches can help us better understand the mechanobiology of EHTs.

O043 Investigating Upper-Airway Mechanics with Computational Tongue Models

Abstract With the use of computational models of the oral cavity, we aim to understand how changes in the morphology of the oral cavity and changes the stiffness of the tongue affect the movement of the posterior surface of the tongue. Anatomical and diffusion weighted imaging from 20 healthy individuals (10M:10F, Age: 22-54years, BMI: 17-30kg/m²), were used to develop 3D models of the tongue including the fibre orientations of the tongue muscles. All models were aligned and scaled to match the volume of the average subject's mandible, so that only differences in shape were modelled. The muscles of the tongue were modelled as transversely isotropic, with a shear moduli of 2.9 and 1.0kPa, parallel and perpendicular to the muscle fibre direction, respectively. Each individual participant model was prescribed an active contraction producing 50kPa of stress in the horizontal component of the genioglossus. The influence of initial hyoid position on the average anterior displacement of the posterior surface of the tongue was assessed with linear regression. On average the contraction moved the posterior tongue 4.0±0.7mm anteriorly. For every millimetre of hyoid elevation, anterior tongue displacement (i.e. airway dilation) decreased by 0.1mm. These initial models demonstrate that the automated modelling pipeline can identify how structural differences alter the efficacy of dilator muscles. Future implementations of these models will also incorporate experimentally measured pharyngeal pressures, and temporal and spatial patterns of muscle activations based on experimental data, to assess the how variation in the coordinated contractions contributes to airway collapse.

Classification of Fracture Risk in Fallers Using Dual-Energy X-Ray Absorptiometry (DXA) Images and Deep Learning-Based Feature Extraction.

Dual-energy X-ray absorptiometry (DXA) scans are one of the most frequently used imaging techniques for calculating bone mineral density, yet calculating fracture risk using DXA image features is rarely performed. The objective of this study was to combine deep neural networks, together with DXA images and patient clinical information, to evaluate fracture risk in a cohort of adults with at least one known fall and age-matched healthy controls. DXA images of the entire body as, well as isolated images of the hip, forearm, and spine (1488 total), were obtained from 478 fallers and 48 non-faller controls. A modeling pipeline was developed for fracture risk prediction using the DXA images and clinical data. First, self-supervised pretraining of feature extractors was performed using a small vision transformer (ViT-S) and a convolutional neural network model (VGG-16 and Resnet-50). After pretraining, the feature extractors were then paired with a multilayer perceptron model, which was used for fracture risk classification. Classification was achieved with an average area under the receiver-operating characteristic curve (AUROC) score of 74.3%. This study demonstrates ViT-S as a promising neural network technique for fracture risk classification using DXA scans. The findings have future application as a fracture risk screening tool for older adults at risk of falls. © 2023 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.

A Fast, Robust Method for Quantitative Assessment of Collagen Fibril Architecture from Transmission Electron Micrographs.

Collagen is the most abundant protein in mammals; it exhibits a hierarchical organization and provides structural support to a wide range of soft tissues, including blood vessels. The architecture of collagen fibrils dictates vascular stiffness and strength, and changes therein can contribute to disease progression. While transmission electron microscopy (TEM) is routinely used to examine collagen fibrils under normal and pathological conditions, computational tools that enable fast and minimally subjective quantitative assessment remain lacking. In the present study, we describe a novel semi-automated image processing and statistical modeling pipeline for segmenting individual collagen fibrils from TEM images and quantifying key metrics of interest, including fibril cross-sectional area and aspect ratio. For validation, we show first-of-their-kind illustrative results for adventitial collagen in the thoracic aorta from three different mouse models.

Enhancing Brain Tumor Classification with Data Augmentation and DenseNet121

This research paper presents a comprehensive study on the development and evaluation of a brain tumor classification model using advanced image processing and deep learning techniques. The primary objective of this study was to create an accurate and robust system for distinguishing between brain tumors and normal brain images, utilizing both an original dataset and an augmented dataset. With a focus on improving medical diagnosis, the research aimed to enhance the performance of brain tumor detection by leveraging state-of-the-art machine learning methods. The model pipeline comprised various image preprocessing steps, including cropping, resizing, denoising, and normalization, followed by feature extraction using the DenseNet121 architecture and classification using sigmoid activation. The dataset was meticulously divided into training, validation, and testing sets, with an emphasis on achieving high recall, precision, F1-score, and accuracy as key research objectives. The results demonstrate that the model achieved impressive performance, with a training recall of 92.87%, precision of 93.82%, F1-score of 93.15%, and an accuracy of 94.83%. These findings underscore the potential of deep learning and data augmentation in enhancing brain tumor detection systems, supporting the research's core objective of advancing medical image analysis for clinical applications.

A Degron Blocking Strategy Towards Improved CRL4CRBN Recruiting PROTAC Selectivity.

Small molecules inducing protein degradation are important pharmacological tools to interrogate complex biology and are rapidly translating into clinical agents. However, to fully realise the potential of these molecules, selectivity remains a limiting challenge. Herein, we addressed the issue of selectivity in the design of CRL4CRBN recruiting PROteolysis TArgeting Chimeras (PROTACs). Thalidomide derivatives used to generate CRL4CRBN recruiting PROTACs have well described intrinsic monovalent degradation profiles by inducing the recruitment of neo-substrates, such as GSPT1, Ikaros and Aiolos. We leveraged structural insights from known CRL4CRBN neo-substrates to attenuate and indeed remove this monovalent degradation function in well-known CRL4CRBN molecular glues degraders, namely CC-885 and Pomalidomide. We then applied these design principles on a previously published BRD9 PROTAC (dBRD9-A) and generated an analogue with improved selectivity profile. Finally, we implemented a computational modelling pipeline to show that our degron blocking design does not impact PROTAC-induced ternary complex formation. We believe that the tools and principles presented in this work will be valuable to support the development of targeted protein degradation.

A taxonomy and review of generalization research in NLP

The ability to generalize well is one of the primary desiderata for models of natural language processing (NLP), but what ‘good generalization’ entails and how it should be evaluated is not well understood. In this Analysis we present a taxonomy for characterizing and understanding generalization research in NLP. The proposed taxonomy is based on an extensive literature review and contains five axes along which generalization studies can differ: their main motivation, the type of generalization they aim to solve, the type of data shift they consider, the source by which this data shift originated, and the locus of the shift within the NLP modelling pipeline. We use our taxonomy to classify over 700 experiments, and we use the results to present an in-depth analysis that maps out the current state of generalization research in NLP and make recommendations for which areas deserve attention in the future.

Optimal Topology of Vision Transformer for Real-Time Video Action Recognition in an End-To-End Cloud Solution

This study introduces an optimal topology of vision transformers for real-time video action recognition in a cloud-based solution. Although model performance is a key criterion for real-time video analysis use cases, inference latency plays a more crucial role in adopting such technology in real-world scenarios. Our objective is to reduce the inference latency of the solution while admissibly maintaining the vision transformer’s performance. Thus, we employed the optimal cloud components as the foundation of our machine learning pipeline and optimized the topology of vision transformers. We utilized UCF101, including more than one million action recognition video clips. The modeling pipeline consists of a preprocessing module to extract frames from video clips, training two-dimensional (2D) vision transformer models, and deep learning baselines. The pipeline also includes a postprocessing step to aggregate the frame-level predictions to generate the video-level predictions at inference. The results demonstrate that our optimal vision transformer model with an input dimension of 56 × 56 × 3 with eight attention heads produces an F1 score of 91.497% for the testing set. The optimized vision transformer reduces the inference latency by 40.70%, measured through a batch-processing approach, with a 55.63% faster training time than the baseline. Lastly, we developed an enhanced skip-frame approach to improve the inference latency by finding an optimal ratio of frames for prediction at inference, where we could further reduce the inference latency by 57.15%. This study reveals that the vision transformer model is highly optimizable for inference latency while maintaining the model performance.

Improved transfer learning using textural features conflation and dynamically fine-tuned layers.

Transfer learning involves using previously learnt knowledge of a model task in addressing another task. However, this process works well when the tasks are closely related. It is, therefore, important to select data points that are closely relevant to the previous task and fine-tune the suitable pre-trained model's layers for effective transfer. This work utilises the least divergent textural features of the target datasets and pre-trained model's layers, minimising the lost knowledge during the transfer learning process. This study extends previous works on selecting data points with good textural features and dynamically selected layers using divergence measures by combining them into one model pipeline. Five pre-trained models are used: ResNet50, DenseNet169, InceptionV3, VGG16 and MobileNetV2 on nine datasets: CIFAR-10, CIFAR-100, MNIST, Fashion-MNIST, Stanford Dogs, Caltech 256, ISIC 2016, ChestX-ray8 and MIT Indoor Scenes. Experimental results show that data points with lower textural feature divergence and layers with more positive weights give better accuracy than other data points and layers. The data points with lower divergence give an average improvement of 3.54% to 6.75%, while the layers improve by 2.42% to 13.04% for the CIFAR-100 dataset. Combining the two methods gives an extra accuracy improvement of 1.56%. This combined approach shows that data points with lower divergence from the source dataset samples can lead to a better adaptation for the target task. The results also demonstrate that selecting layers with more positive weights reduces instances of trial and error in selecting fine-tuning layers for pre-trained models.

Joint UV Optimization and Texture Baking

Level of detail has been widely used in interactive computer graphics. In current industrial 3D modeling pipelines, artists rely on commercial software to generate highly detailed models with UV maps and then bake textures for low-poly counterparts. In these pipelines, each step is performed separately, leading to unsatisfactory visual appearances for low polygon count models. Moreover, existing texture baking techniques assume the low-poly mesh has a small geometric difference from the high-poly, which is often not true in practice, especially with extremely low poly count models. To alleviate the visual discrepancy of the low-poly mesh, we propose to jointly optimize UV mappings during texture baking, allowing for low-poly models to faithfully replicate the appearance of the high-poly even with large geometric differences. We formulate the optimization within a differentiable rendering framework, allowing the automatic adjustment of texture regions to encode appearance information. To compensate for view parallax when two meshes have large geometric differences, we introduce a spherical harmonic parallax mapping, which uses spherical harmonic functions to modulate per-texel UV coordinates based on the view direction. We evaluate the effectiveness and robustness of our approach on a dataset composed of online downloaded models, with varying complexities and geometric discrepancies. Our method achieves superior quality over state-of-the-art techniques and commercial solutions.

Automated measurements of interscrew angles in vertebral body tethering patients with deep learning

BACKGROUND CONTEXTVertebral body tethering is the most popular nonfusion treatment for adolescent idiopathic scoliosis. The effect of the tether cord on the spine can be segmentally assessed by comparing the angle between two adjacent screws (interscrew angle) over time. Tether breakage has historically been assessed radiographically by a change in adjacent interscrew angle by greater than 5° between two sets of imaging. A threshold for growth modulation has not yet been established in the literature. These angle measurements are time consuming and prone to interobserver variability. PURPOSEThe purpose of this study was to develop an automated deep learning algorithm for measuring the interscrew angle following VBT surgery. STUDY DESIGN/SETTINGSingle institution analysis of medical images. PATIENT SAMPLEWe analyzed 229 standing or bending AP or PA radiographs from 100 patients who had undergone VBT at our institution. OUTCOME MEASURESPhysiologic Measures: An image processing algorithm was used to measure interscrew angles. METHODSA total of 229 standing or bending AP or PA radiographs from 100 VBT patients with vertebral body tethers were identified. Vertebral body screws were segmented by hand for all images and interscrew angles measured manually for 60 of the included images. A U-Net deep learning model was developed to automatically segment the vertebral body screws. Screw label maps were used to develop and tune an image processing algorithm which measures interscrew angles. Finally, the completed model and algorithm pipeline was tested on a 30-image test set. Dice score and absolute error were used to measure performance. RESULTSInter- and Intra-rater reliability for manual angle measurements were assessed with ICC and were both 0.99. The segmentation model Dice score against manually segmented ground truth across the 30-image test set was 0.96. The average interscrew angle absolute error between the algorithm and manually measured ground truth was 0.66° and ranged from 0° to 2.67° in non-overlapping screws (N=206). The primary modes of failure for the model were overlapping screws on a right thoracic/left lumbar construct with two screws in one vertebra and overexposed images. An algorithm step which determines whether an overlapping screw was present correctly identified all overlapping screws, with no false positives. CONCLUSIONWe developed and validated an algorithm which measures interscrew angles for radiographs of vertebral body tether patients with an accuracy of within 1° for the majority of interscrew angles. The algorithm can process five images per second on a standard computer, leading to substantial time savings. This algorithm may be used for rapid processing of large radiographic databases of tether patients and could enable more rigorous definitions of growth modulation and cord breakage to be established.

A Rapid-Patterning 3D Vessel-on-Chip for Imaging and Quantitatively Analyzing Cell-Cell Junction Phenotypes.

The blood-brain barrier (BBB) is a dynamic interface that regulates the molecular exchanges between the brain and peripheral blood. The permeability of the BBB is primarily regulated by the junction proteins on the brain endothelial cells. In vitro BBB models have shown great potential for the investigation of the mechanisms of physiological function, pathologies, and drug delivery in the brain. However, few studies have demonstrated the ability to monitor and evaluate the barrier integrity by quantitatively analyzing the junction presentation in 3D microvessels. This study aimed to fabricate a simple vessel-on-chip, which allows for a rigorous quantitative investigation of junction presentation in 3D microvessels. To this end, we developed a rapid protocol that creates 3D microvessels with polydimethylsiloxane and microneedles. We established a simple vessel-on-chip model lined with human iPSC-derived brain microvascular endothelial-like cells (iBMEC-like cells). The 3D image of the vessel structure can then be "unwrapped" and converted to 2D images for quantitative analysis of cell-cell junction phenotypes. Our findings revealed that 3D cylindrical structures altered the phenotype of tight junction proteins, along with the morphology of cells. Additionally, the cell-cell junction integrity in our 3D models was disrupted by the tumor necrosis factor α. This work presents a "quick and easy" 3D vessel-on-chip model and analysis pipeline, together allowing for the capability of screening and evaluating the cell-cell junction integrity of endothelial cells under various microenvironment conditions and treatments.

Prognostic model of kidney renal clear cell carcinoma using aging-related long noncoding RNA signatures identifies THBS1-IT1 as a potential prognostic biomarker for multiple cancers.

Aging is responsible for the main intrinsic triggers of cancers; however, the studies of aging risk factors in cancer animal models and cancer patients are rare and insufficient to be represented in cancer clinical trials. For a better understanding of the complex regulatory networks of aging and cancers, 8 candidate aging related long noncoding RNAs (CarLncs) identified from the healthy aging models, centenarians and their offsprings, were selected and their association with kidney renal clear cell carcinoma (KIRC) was explored by series of cutting edge analyses such as support vector machine (SVM) and random forest (RF) algorithms. Using data downloaded from TCGA and GTEx databases, a regulatory network of CarLncs-miRNA-mRNA was constructed and five genes within the network were screened out as aging related feature genes for developing KIRC prognostic models. After a strict filtering pipeline for modeling, a formula using the transcript per million (TPM) values of feature genes "LncAging_score = 0.008* MMP11 + 0.066* THBS1-IT1 + (-0.014)* DYNLL2 + (-0.030)* RMND5A+ 0.008* PEG10" was developed. ROC analysis and nomogram suggest our model achieves a great performance in KIRC prognosis. Among the 8 CarLncs, we found that THBS1-IT1 was significantly dysregulated in 12 cancer types. A comprehensive pan-cancer analysis demonstrated that THBS1-IT1 is a potential prognostic biomarker in not only KIRC but also multiple cancers, such as LUSC, BLCA, GBM, LGG, MESO, PAAD, STAD and THCA, it was correlated with tumor microenvironment (TME) and tumor immune cell infiltration (TICI) and its high expression was related with poor survival.

A Statistical Shape Model Pipeline to Enable the Creation of Synthetic 3D Liver Data

Abstract The application of machine learning approaches in medical technology is gaining more and more attention. Due to the high restrictions for collecting intraoperative patient data, synthetic data is increasingly used to support the training of artificial neural networks. We present a pipeline to create a statistical shape model (SSM) using 28 segmented clinical liver CT scans. Our pipeline consists of four steps: data preprocessing, rigid alignment, template morphing, and statistical modeling. We compared two different template morphing approaches: Laplace-Beltrami-regularized projection (LBRP) and nonrigid iterative closest points translational (N-ICP-T) and evaluated both morphing approaches and their corresponding shape model performance using six metrics. LBRP achieved a smaller mean vertex-to-nearest-neighbor distances (2.486±0.897 mm) than N-ICP-T (5.559±2.413 mm). Generalization and specificity errors for LBRP were consistently lower than those of N-ICP-T. The first principal components of the SSM showed realistic anatomical variations. The performance of the SSM was comparable to a state-of-the-art model.

AIWFF: A Machine Learning based Framework for Automatic Weather Forecasting

In the contemporary era witnessing global warming effects, weather is a dynamic phenomenon which is highly uncertain. The conventional approaches that rely on certain physical processes governing atmosphere are capable of serving a great deal in weather forecasting. However, robustness to perturbations is still desired. In this content Artificial Intelligence (AI) innovations assume significance to bring about more reliable forecasting alternative which may complement conventional methods. In this paper, we proposed a framework known as AI-enabled Weather Forecasting Framework (AIWFF) which exploits machine learning (ML) models that are robust to time series data and underlying perturbations for improving forecasting performance. An algorithm known as Learning based Intelligent Weather Forecasting (LIWF) is proposed and implemented. This algorithm has required pre-processing, feature section and a pipeline of ML models to learn from data and then forecast weather more accurately. Another algorithm known as Hybrid Method for Feature Selection (HMFS) is proposed to leverage training quality in LIWF algorithm. The framework results in three trained knowledge models saved to secondary storage. These models are known as Random Forest Regressor, Linear Regressor and Decision Tree Regressor. An application with Graphical User Interface (GUI) is developed to make use of these knowledge models and provide forecasting on user requests. The empirical results revealed that the proposed framework shows better performance.

Artificial intelligence for dementia research methods optimization.

Artificial intelligence (AI) and machine learning (ML) approaches are increasingly being used in dementia research. However, several methodological challenges exist that may limit the insights we can obtain from high-dimensional data and our ability to translate these findings into improved patient outcomes. To improve reproducibility and replicability, researchers should make their well-documented code and modeling pipelines openly available. Data should also be shared where appropriate. To enhance the acceptability of models and AI-enabled systems to users, researchers should prioritize interpretable methods that provide insights into how decisions are generated. Models should be developed using multiple, diverse datasets to improve robustness, generalizability, and reduce potentially harmful bias. To improve clarity and reproducibility, researchers should adhere to reporting guidelines that are co-produced with multiple stakeholders. If these methodological challenges are overcome, AI and ML hold enormous promise for changing the landscape of dementia research and care. HIGHLIGHTS: Machine learning (ML) can improve diagnosis, prevention, and management of dementia. Inadequate reporting of ML procedures affects reproduction/replication of results. ML models built on unrepresentative datasets do not generalize to new datasets. Obligatory metrics for certain model structures and use cases have not been defined. Interpretability and trust in ML predictions are barriers to clinical translation.