Symbolic regression of generative network models.
Ontology highlight
ABSTRACT:
SUBMITTER: Menezes T
PROVIDER: S-EPMC4155339 | biostudies-literature | 2014 Sep
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Machine learning (ML) models trained for triggering clinical decision support (CDS) are typically either accurate or interpretable but not both. Scaling CDS to the panoply of clinical use cases while mitigating risks to patients will require many ML models be intuitively interpretable for clinicians. To this end, we adapted a symbolic regression method, coined the feature engineering automation tool (FEAT), to train concise and accurate models from high-dimensional electronic health record (EHR) data. We first present an in-depth application of FEAT to classify hypertension, hypertension with unexplained hypokalemia, and apparent treatment-resistant hypertension (aTRH) using EHR data for 1200 subjects receiving longitudinal care in a large healthcare system. FEAT models trained to predict phenotypes adjudicated by chart review had equivalent or higher discriminative performance (p < 0.001) and were at least three times smaller (p < 1 × 10-6) than other potentially interpretable models. For aTRH, FEAT generated a six-feature, highly discriminative (positive predictive value = 0.70, sensitivity = 0.62), and clinically intuitive model. To assess the generalizability of the approach, we tested FEAT on 25 benchmark clinical phenotyping tasks using the MIMIC-III critical care database. Under comparable dimensionality constraints, FEAT's models exhibited higher area under the receiver-operating curve scores than penalized linear models across tasks (p < 6 × 10-6). In summary, FEAT can train EHR prediction models that are both intuitively interpretable and accurate, which should facilitate safe and effective scaling of ML-triggered CDS to the panoply of potential clinical use cases and healthcare practices.
| S-EPMC10241925 | biostudies-literature
Project description:Estimating parameters for distributed hydrological models is a challenging and long studied task. Parameter transfer functions, which define model parameters as functions of geophysical properties of a catchment, might improve the calibration procedure, increase process realism, and can enable prediction in ungauged areas. We present the function space optimization (FSO), a symbolic regression method for estimating parameter transfer functions for distributed hydrological models. FSO is based on the idea of transferring the search for mathematical expressions into a continuous vector space that can be used for optimization. This is accomplished by using a text generating neural network with a variational autoencoder architecture that can learn to compress the information of mathematical functions. To evaluate the performance of FSO, we conducted a case study using a parsimonious hydrological model and synthetic discharge data. The case study consisted of two FSO applications: single-criteria FSO, where only discharge was used for optimization, and multicriteria FSO, where additional spatiotemporal observations of model states were used for transfer function estimation. The results show that FSO is able to estimate transfer functions correctly or approximate them sufficiently. We observed a reduced fit of the parameter density functions resulting from the inferred transfer functions for less sensitive model parameters. For those it was sufficient to estimate functions resulting in parameter distributions with approximately the same mean parameter values as the real transfer functions. The results of the multicriteria FSO showed that using multiple spatiotemporal observations for optimization increased the quality of estimation considerably.
| S-EPMC7583385 | biostudies-literature
Project description:Understanding the dynamics of complex ecosystems is a necessary step to maintain and control them. Yet, reverse-engineering ecological dynamics remains challenging largely due to the very broad class of dynamics that ecosystems may take. Here, this challenge is tackled through symbolic regression, a machine learning method that automatically reverse-engineers both the model structure and parameters from temporal data. How combining symbolic regression with a "dictionary" of possible ecological functional responses opens the door to correctly reverse-engineering ecosystem dynamics, even in the case of poorly informative data, is shown. This strategy is validated using both synthetic and experimental data, and it is found that this strategy is promising for the systematic modeling of complex ecological systems.
| S-EPMC7339472 | biostudies-literature
Project description:Eye tracking systems have recently experienced a diversity of novel calibration procedures, including smooth pursuit and vestibulo-ocular reflex based calibrations. These approaches allowed collecting more data compared to the standard 9-point calibration. However, the computation of the mapping function which provides planar gaze positions from pupil features given as input is mostly based on polynomial regressions, and little work has investigated alternative approaches. This paper fills this gap by providing a new calibration computation method based on symbolic regression. Instead of making prior assumptions on the polynomial transfer function between input and output records, symbolic regression seeks an optimal model among different types of functions and their combinations. This approach offers an interesting perspective in terms of flexibility and accuracy. Therefore, we designed two experiments in which we collected ground truth data to compare vestibulo-ocular and smooth pursuit calibrations based on symbolic regression, both using a marker or a finger as a target, resulting in four different calibrations. As a result, we improved calibration accuracy by more than 30%, with reasonable extra computation time.
| S-EPMC6420251 | biostudies-literature
Project description:Symbolic regression (SR) is a machine learning-based regression method based on genetic programming principles that integrates techniques and processes from heterogeneous scientific fields and is capable of providing analytical equations purely from data. This remarkable characteristic diminishes the need to incorporate prior knowledge about the investigated system. SR can spot profound and elucidate ambiguous relations that can be generalizable, applicable, explainable and span over most scientific, technological, economical, and social principles. In this review, current state of the art is documented, technical and physical characteristics of SR are presented, the available programming techniques are investigated, fields of application are explored, and future perspectives are discussed. Supplementary Information The online version contains supplementary material available at 10.1007/s11831-023-09922-z.
| S-EPMC10113133 | biostudies-literature
Project description:PurposeTo evaluate pix2pix and CycleGAN and to assess the effects of multiple combination strategies on accuracy for patch-based synthetic computed tomography (sCT) generation for magnetic resonance (MR)-only treatment planning in head and neck (HN) cancer patients.Materials and methodsTwenty-three deformably registered pairs of CT and mDixon FFE MR datasets from HN cancer patients treated at our institution were retrospectively analyzed to evaluate patch-based sCT accuracy via the pix2pix and CycleGAN models. To test effects of overlapping sCT patches on estimations, we (a) trained the models for three orthogonal views to observe the effects of spatial context, (b) we increased effective set size by using per-epoch data augmentation, and (c) we evaluated the performance of three different approaches for combining overlapping Hounsfield unit (HU) estimations for varied patch overlap parameters. Twelve of twenty-three cases corresponded to a curated dataset previously used for atlas-based sCT generation and were used for training with leave-two-out cross-validation. Eight cases were used for independent testing and included previously unseen image features such as fused vertebrae, a small protruding bone, and tumors large enough to deform normal body contours. We analyzed the impact of MR image preprocessing including histogram standardization and intensity clipping on sCT generation accuracy. Effects of mDixon contrast (in-phase vs water) differences were tested with three additional cases. The sCT generation accuracy was evaluated using mean absolute error (MAE) and mean error (ME) in HU between the plan CT and sCT images. Dosimetric accuracy was evaluated for all clinically relevant structures in the independent testing set and digitally reconstructed radiographs (DRRs) were evaluated with respect to the plan CT images.ResultsThe cross-validated MAEs for the whole-HN region using pix2pix and CycleGAN were 66.9 ± 7.3 vs 82.3 ± 6.4 HU, respectively. On the independent testing set with additional artifacts and previously unseen image features, whole-HN region MAEs were 94.0 ± 10.6 and 102.9 ± 14.7 HU for pix2pix and CycleGAN, respectively. For patients with different tissue contrast (water mDixon MR images), the MAEs increased to 122.1 ± 6.3 and 132.8 ± 5.5 HU for pix2pix and CycleGAN, respectively. Our results suggest that combining overlapping sCT estimations at each voxel reduced both MAE and ME compared to single-view non-overlapping patch results. Absolute percent mean/max dose errors were 2% or less for the PTV and all clinically relevant structures in our independent testing set, including structures with image artifacts. Quantitative DRR comparison between planning CTs and sCTs showed agreement of bony region positions to <1 mm.ConclusionsThe dosimetric and MAE based accuracy, along with the similarity between DRRs from sCTs, indicate that pix2pix and CycleGAN are promising methods for MR-only treatment planning for HN cancer. Our methods investigated for overlapping patch-based HU estimations also indicate that combining transformation estimations of overlapping patches is a potential method to reduce generation errors while also providing a tool to potentially estimate the MR to CT aleatoric model transformation uncertainty. However, because of small patient sample sizes, further studies are required.
| S-EPMC7146715 | biostudies-literature
Project description:When individuals interact with each other and meaningfully contribute toward a common goal, it results in a collaboration. The artifacts resulting from collaborations are best captured using a hypergraph model, whereas the relation of who has collaborated with whom is best captured via an abstract simplicial complex (SC). We propose a generative algorithm GENESCs for SCs modeling fundamental collaboration relations. The proposed network growth process favors attachment that is preferential not to an individual's degree, i.e., how many people has he/she collaborated with, but to his/her facet degree, i.e., how many maximal groups or facets has he/she collaborated within. Based on our observation that several real-world facet size distributions have significant deviation from power law-mainly since larger facets tend to subsume smaller ones-we adopt a data-driven approach. We prove that the facet degree distribution yielded by GENESCs is power law distributed for large SCs and show that it is in agreement with real world co-authorship data. Finally, based on our intuition of collaboration formation in domains such as collaborative scientific experiments and movie production, we propose two variants of GENESCs based on clamped and hybrid preferential attachment schemes, and show that they perform well in these domains.
| S-EPMC5593992 | biostudies-literature
Project description: Not available
| S-EPMC11339076 | biostudies-literature
Project description:Studies of brain activity during number processing suggest symbolic and nonsymbolic numerical stimuli (e.g., Arabic digits and dot arrays) engage both shared and distinct neural mechanisms. However, the extent to which number format influences large-scale functional network organization is unknown. In this study, using 7 Tesla MRI, we adopted a network neuroscience approach to characterize the whole-brain functional architecture supporting symbolic and nonsymbolic number comparison in 33 adults. Results showed the degree of global modularity was similar for both formats. The symbolic format, however, elicited stronger community membership among auditory regions, whereas for nonsymbolic, stronger membership was observed within and between cingulo-opercular/salience network and basal ganglia communities. The right posterior inferior temporal gyrus, left intraparietal sulcus, and two regions in the right ventromedial occipital cortex demonstrated robust differences between formats in terms of their community membership, supporting prior findings that these areas are differentially engaged based on number format. Furthermore, a unified fronto-parietal/dorsal attention community in the nonsymbolic condition was fractionated into two components in the symbolic condition. Taken together, these results reveal a pattern of overlapping and distinct network architectures for symbolic and nonsymbolic number processing.
| S-EPMC7462424 | biostudies-literature
Project description:We develop and apply analytically tractable generative models of livestock movements at national scale. These go beyond current models through mechanistic modelling of heterogeneous trade partnership network dynamics and the trade events that occur on them. Linking resulting animal movements to disease transmission between farms yields analytical expressions for the basic reproduction number R 0. We show how these novel modelling tools enable systems approaches to disease control, using R 0 to explore impacts of changes in trading practices on between-farm prevalence levels. Using the Scottish cattle trade network as a case study, we show our approach captures critical complexities of real-world trade networks at the national scale for a broad range of endemic diseases. Changes in trading patterns that minimize disruption to business by maintaining in-flow of animals for each individual farm reduce R 0, with the largest reductions for diseases that are most challenging to eradicate. Incentivizing high-risk farms to adopt such changes exploits 'scale-free' properties of the system and is likely to be particularly effective in reducing national livestock disease burden and incursion risk. Encouragingly, gains made by such targeted modification of trade practices scale much more favourably than comparably targeted improvements to more commonly adopted farm-level biosecurity.
| S-EPMC8074963 | biostudies-literature