Voxel-based Features Research Articles

Manufacturing Feature Recognition with a Sparse Voxel-based Convolutional Neural Network

Abstract Automated manufacturing feature recognition is a crucial link between computer-aided design and manufacturing, facilitating process selection and other downstream tasks in computer-aided process planning. While various methods such as graph-based, rule-based, and neural networks have been proposed for automatic feature recognition, they suffer from poor scalability or computational inefficiency. Recently, voxel-based convolutional neural networks have shown promise in solving these challenges but incur a tradeoff between computational cost and feature resolution. This paper investigates a computationally efficient sparse voxel-based convolutional neural network for manufacturing feature recognition, specifically, an octree-based sparse voxel convolutional neural network. This model is trained on a large-scale manufacturing feature dataset, and its performance is compared to a voxel-based feature recognition model (FeatureNet). The results indicate that the octree-based model yields higher feature recognition accuracy (99.5% on the test dataset) with 44% lower GPU memory consumption than a voxel-based model of comparable resolution. In addition, increasing the resolution of the octree-based model enables recognition of finer manufacturing features. These results indicate that a sparse voxel-based convolutional neural network is a computationally efficient deep learning model for manufacturing feature recognition to enable process planning automation. Moreover, the sparse voxel-based neural network demonstrated comparable performance to a boundary representation-based feature recognition neural network, achieving similar accuracy in single feature recognition without having access to the exact 3D shape descriptors

PSVMLP: Point and Shifted Voxel MLP for 3D deep learning

We propose a high-performance 3D feature extraction deep learning network based on point cloud and shifted voxel, named Point and Shifted Voxel MLP (PSVMLP). The main component of PSVMLP is simple Multi-Layer Perceptron (MLP) structure. PSVMLP achieves effective extraction of multi-scale features from 3D data. Specifically, we combine point cloud and voxel-based feature extraction methods. In voxel representation learning, we propose a wide-range geometric feature extraction method based on axial shifting operations and simple MLP structure. The axial shifting operations allow shifting voxels in the depth, height, and width directions, capturing more geometric information. In point cloud representation learning, we use simple MLP structure to extract local features, and we also extract global features by combining transformer structure. By combining point cloud and voxel feature extraction methods, we obtain rich feature representations from different scales, enhancing the model’s expressive power and generalization performance. Applying our designed model to basic geometric feature learning tasks, we achieve excellent results. Despite being built primarily on a simple MLP framework, our model demonstrates remarkable performance on both shape classification and shape part segmentation tasks. Our code is available at https://github.com/hitxraz/psvmlp.

Development of a High-Precision Lidar System and Improvement of Key Steps for Railway Obstacle Detection Algorithm

In response to the growing demand for railway obstacle monitoring, lidar technology has emerged as an up-and-coming solution. In this study, we developed a mechanical 3D lidar system and meticulously calibrated the point cloud transformation to monitor specific areas precisely. Based on this foundation, we have devised a novel set of algorithms for obstacle detection within point clouds. These algorithms encompass three key steps: (a) the segmentation of ground point clouds and extraction of track point clouds using our RS-Lo-RANSAC (region select Lo-RANSAC) algorithm; (b) the registration of the BP (background point cloud) and FP (foreground point cloud) via an improved Robust ICP algorithm; and (c) obstacle recognition based on the VFOR (voxel-based feature obstacle recognition) algorithm from the fused point clouds. This set of algorithms has demonstrated robustness and operational efficiency in our experiments on a dataset obtained from an experimental field. Notably, it enables monitoring obstacles with dimensions of 15 cm × 15 cm × 15 cm. Overall, our study showcases the immense potential of lidar technology in railway obstacle monitoring, presenting a promising solution to enhance safety in this field.

Accelerating Point-Voxel Representation of 3-D Object Detection for Automatic Driving

Current point-voxel fusion methods for 3D object detection could not make full use of complementary information in the field of autonomous driving. Therefore, a novel two-stage 3D object detection method, called Accelerating Point-Voxel Representation (APVR), is proposed. The advantages of Point-based feature and Voxel-based feature can be integrated into a single 3D representation. Thereby, the proposed method retains more fine-grained information of an object while maintaining high efficiency. Specifically, computational cost is reduced by adding offsets to query neighboring voxels of key-points. More fine-grained information can be obtained by calculating the matching probability between neighbouring voxels and key-points. During the optimization of the prediction boxes, virtual grid points are set to capture the spatial information between key-points. The constraint of minimum enclosing rectangle is also added to optimize the directions of the prediction boxes. A large number of experiments on the KITTI, NuScenes and Waymo datasets demonstrate great generalizability and portability of the proposed approach. The effectiveness and efficiency of APVR has been proved by comparisons with the state-of-art methods. APVR makes the real-time processing frame rate reach 40.4 Hz while ensuring high prediction accuracy.

Stem Detection from Terrestrial Laser Scanning Data with Features Selected via Stem-Based Evaluation

Terrestrial laser scanning (TLS) is an effective tool for extracting stem distribution, providing essential information for forest inventory and ecological studies while also assisting forest managers in monitoring and controlling forest stand density. A feature-based method is commonly integrated into the pipelines of stem detection, facilitating the transition from stem point to stem instance, but most studies focus on feature effectiveness from the point level, neglecting the relationship between stem point extraction and stem detection. In this paper, a feature-based method is proposed to identify stems from TLS data, with features selected from stem levels. Firstly, we propose a series of voxel-based features considering the stem characteristics under the forest. Then, based on the evaluation of some commonly used and proposed features, a stem-based feature selection method is proposed to select a suitable feature combination for stem detection by constructing and evaluating different combinations. Experiments are carried out on three plots with different terrain slopes and tree characteristics, each having a sample plot size of about 8000 m2. The results show that the voxel-based features can supplement the basic features, which improve the average accuracy of stem point extraction and stem detection by 9.5% and 1.2%, respectively. The feature set obtained by the proposed feature selection method achieves a better balance between accuracy and feature number compared with the point-based feature selection method and the features used in previous studies. Moreover, the accuracies of the proposed stem detection methods are also comparable to the three methods evaluated in the international TLS benchmarking project.

A neuromarker for deficit syndrome in schizophrenia from a combination of structural and functional magnetic resonance imaging.

Deficit schizophrenia (DS), defined by primary and enduring negative symptoms, has been proposed as a promising homogeneous subtype of schizophrenia. It has been demonstrated that unimodal neuroimaging characteristics of DS were different from non-deficit schizophrenia (NDS), however, whether multimodal-based neuroimaging features could identify deficit syndrome remains to be determined. Functional and structural multimodal magnetic resonance imaging of DS, NDS and healthy controls were scanned. Voxel-based features of gray matter volume, fractional amplitude of low-frequency fluctuations, and regional homogeneity were extracted. The support vector machine classification models were constructed using these features separately and jointly. The most discriminative features were defined as the first 10% of features with the greatest weights. Moreover, relevance vector regression was applied to explore the predictive values of these top-weighted features in predicting negative symptoms. The multimodal classifier achieved a higher accuracy (75.48%) compared with the single modal model in distinguishing DS from NDS. The most predictive brain regions were mainly located in the default mode and visual networks, exhibiting differences between functional and structural features. Further, the identified discriminative features significantly predicted scores of diminished expressivity factor in DS but not NDS. The present study demonstrated that local properties of brain regions extracted from multimodal imaging data could distinguish DS from NDS with a machine learning-based approach and confirmed the relationship between distinctive features and the negative symptoms subdomain. These findings may improve the identification of potential neuroimaging signatures and improve the clinical assessment of the deficit syndrome.

Multi Point-Voxel Convolution (MPVConv) for deep learning on point clouds

The existing 3D deep learning methods adopt either individual point-based features or local-neighboring voxel-based features, and demonstrate great potential for processing 3D data. However, the point-based models are inefficient due to the unordered nature of point clouds and the voxel-based models suffer from large information loss. Motivated by the success of recent point-voxel representation, such as PVCNN and DRINet, we propose a new convolutional neural network, called Multi Point-Voxel Convolution (MPVConv), for deep learning on point clouds. Integrating both the advantages of voxel and point-based methods, MPVConv can effectively increase the neighboring collection between point-based features and also promote independence among voxel-based features. Extensive experiments on benchmark datasets such as ShapeNet Part, S3DIS and KITTI for various tasks show that MPVConv improves the accuracy of the backbone (PointNet) by up to 36%, and achieves higher accuracy than the voxel-based model with up to 34× speedups. In addition, MPVConv outperforms the state-of-the-art point-based models with up to 8× speedups. Also, our MPVConv only needs 65% of the GPU memory required by the latest point-voxel-based model (DRINet). The source code of our method is attached in https://github.com/NWUzhouwei/MPVConv.

NIMG-73. PREDICTING PERITUMORAL GLIOBLASTOMA INFILTRATION USING QUANTITATIVE MAGNETIC RESONANCE FINGERPRINTING RADIOMICS

Abstract PURPOSE Standard clinical magnetic resonance (MR) imaging of glioblastoma (GB) cannot identify malignant infiltration into the peritumoral non-enhancing region. This work implements quantitative MR fingerprinting (MRF) based radiomics to predict infiltrative regions of GB. METHODS Pre-operative MRF T1 and T2 maps along with multiparametric MR (mpMR) images (T1w, T2w, T1w-Gd, FLAIR, and ADC) from GB patients (n = 10) were analyzed. All subjects had histologically confirmed sites of non-enhancing peritumoral infiltration as identified by intra-operative 5-ALA fluorescence-guided tissue resection. These locations (n = 40) were manually annotated on pre-operative FLAIR by a board-certified neuroradiologist and labeled as “infiltration” (INF). For each patient, another FLAIR region ( > 3 cm from the enhancing tumor margin) was identified and labeled as “edema” (ED). Following image co-registration, 693 handcrafted radiomic voxel-based features were extracted from confirmed infiltration (INF) and edema (ED) voxels using a 3D 5x5x5 voxel sliding kernel. Feature selection was performed using the minimum redundancy maximal relevance (MRMR) algorithm, then INF and ED voxel features were used to train two cross-validated binary support vector machine (SVM) models for voxel-wise infiltration prediction: 1) MRF only model and 2) combined MRF and mpMRI model. Following MRMR selection, the combined model included features from T1, T1w, T2w, T1w-Gd, FLAIR, and ADC images. Model performance was evaluated using one withheld subject with multiple known infiltration sites. Balanced test accuracy and receiver operating characteristic (ROC) curve analysis was used to evaluate classification performance. RESULTS The MRF only and combined model achieved balanced test accuracies of 77.87% and 73.21% and AUCs of 0.91 and 0.97, respectively. CONCLUSIONS This study demonstrates that MRF-based radiomic features can predict GB tumor infiltration with high accuracy and provide complementary value to standard clinical MRI features. Our results indicate the high potential of employing MRF to guide personalized GB treatment strategies.

Camera and LiDAR-based point painted voxel region-based convolutional neural network for robust 3D object detection

Most of the three-dimensional (3D) object detection methods based on LiDAR point cloud data achieve relatively high performance in general cases. However, when the LiDAR points have noise or some corruptions, the detection performance can be severely affected. We propose a 3D object detection method that combines point cloud information with two-dimensional (2D) semantic segmentation information to enhance the feature representation for difficult cases, such as sparse, noisy, and partially absent data. Motivated by the Pointpainting techniques, we designed an early-stage fusion method based on a Voxel region-based convolutional neural network (R-CNN) architecture. The 2D semantic segmentation scores obtained by the Pointpainting techniques are appended to the raw point cloud data. The voxel-based features and 2D semantic information improve the performance in detecting instances when the point cloud is corrupted. In addition, we also designed a multiscale hierarchical region of interest pooling strategy that reduced the computational cost of Voxel R-CNN by at least 43%. Our method shows competitive results with the state-of-the-art methods on the standard KITTI dataset. In addition, three corrupted KITTI datasets, KITTI sparse (KITTI-S), KITTI jittering (KITTI-J), and KITTI dropout (KITTI-D), were used for robustness testing. With the noisy LiDAR points, our proposed point painted Voxel R-CNN achieved superior detection performance over that of the baseline Voxel R-CNN for the moderate case, with a notable improvement of 11.13% in average precision (AP) on the 3D object detection and 14.3% in AP on the bird’s eye view object detection.

Associations Of Visceral Adipose Tissue With Chronological And BrainAge

New analytic approaches that leverage machine learning have been developed to assess biological brain age (BrainAge), as opposed to chronological age. The difference between chronological age and BrainAge (Brain Predicted Age Difference [BPAD]) may be an important health indicator, which invites exploration of factors associated with variation in BPAD. While adipose tissue in general, and visceral adipose tissue (VAT) in particular, has been associated with numerous chronic diseases and adverse outcomes in some specific brain structures, to our knowledge there has been no exploration of the relationship between VAT and BPAD. PURPOSE: To explore the associations of chronological age, BrainAge, and BPAD with VAT in older adults (65-85 yrs). METHODS: Healthy women (n = 404) and men (n = 149), mean age 72 ± 5 yrs, were assessed for VAT by Dual X-Ray Absorptiometry and received a structural MRI scan processed using the BrainAgeR algorithm developed by Cole et al (2018) using voxel-based features from a training sample of adults (n = 3377) aged 18-92, to derive BrainAge. Pearson correlation was used to assess associations between VAT and both chronological age and BrainAge, and linear regression was utilized to further assess the association between VAT and BPAD, adjusted for chronological age, sex, and physical activity. RESULTS: VAT was significantly associated with BrainAge (r(551) = 0.133 p < 0.002) but not chronological age (r(551) = 0.004 p = 0.460). VAT significantly predicted BPAD after adjusting for sex, chronological age, and physical activity (p = .001) with each unit of increase of VAT predicting a 0.99 year difference in BPAD (indicating an older than chronological age brain) as shown in Table 1.CONCLUSION: Higher VAT mass was associated with more advanced brain age as determined by the BrainAgeR technique. Future research should determine if changes in VAT deposition may positively affect brain structure in older adults.

SMS-Net: Sparse multi-scale voxel feature aggregation network for LiDAR-based 3D object detection

Real-time three-dimensional (3D) object detection has become a crucial component of autonomous driving applications. Recent research demonstrates that a voxel-based feature aggregation method is accurate and efficient in large 3D scenes. However, the parameter choice of voxel size has become a sensitive issue because of the contradiction between its detection performance and inference speed. To alleviate this problem, in this paper we propose a sparse multi-scale voxel feature aggregation network (SMS-Net), a novel one-stage, end-to-end network that primarily contains a sparse multi-scale-fusion (SMSF) module and shallow-to-deep regression (SDR) module. First, the raw point clouds are divided into different scales of voxels to construct diverse 3D sparse feature maps. Then, the SMSF module attentively aggregates the point-wise features with a perspective-channel attention mechanism and fuses multi-scale features at 3D sparse feature-map level to achieve more fine-grained shape information. In addition, the new SDR module boosts the localization accuracy and 3D box estimation accuracy through multiple aggregation at feature-map level, which requires less computational overhead. Extensive experiments demonstrate the remarkable performance improvements from each module of the proposed method. On the KITTI 3D object detection benchmark, for example, SMS-Net outperforms most one-stage, state-of-the-art methods and its performance can even be compared to that of two-stage methods. These detection results are achieved with a real-time inference speed of 42 Hz. SMS-Net also achieves state-of-the-art performance on the nuScenes 3D benchmark.

Voxel-based radiomics outlines spatial heterogeneity of cerebral radiation necrosis (RN) associated with bevacizumab (Bev) response in head and neck radiotherapy (RT) patients.

e18063 Background: RN is a debilitating toxicity of head and neck RT, with incidences of 4% to 24%. Bev is a promising drug for managing RN, but there is currently no biomarker that could predict a priori if a patient would benefit from Bev. Here, we aimed to use radiomics to predict Bev response. However, current radiomics tools suffer from issues of interrater contouring variability and explanability of the radiomics models. We therefore investigated a voxel-based radiomics approach to characterize the spatial heterogeneity of RN, and define subregions that are associated with Bev response. Methods: 118 consecutive nasopharyngeal carcinoma (NPC) patients who were diagnosed with RN post-RT, and were treated with Bev between Jul 2012 and Mar 2019 at the Sun Yat-Sen Memorial Hospital were enrolled into the study. 77 patients with 101 brain lesions treated between Jul 2012 and Dec 2017 were assigned to the training set, and 41 patients with 51 brain lesions treated between Jan 2018 and Mar 2019 were used for validation. All patients received Bev prescribed at 5 mg/kg every fortnightly for up to 4 courses. We extracted voxel-based radiomics features from each segment drawn on T2 FLAIR MRI images, followed by a 3-step analysis (individual- and population-level clustering, before delta-radiomics) to obtain a clustered heat map of subregions within the RN lesion. Anatomical correlation to areas of edema, necrosis, and their interphase was performed. Delta-radiomics tracked the ∆ of each cluster to Bev. Features were then extracted from each radiomics cluster for model building. Results: 71 (70.3%) and 34 (66.7%) lesions had documented radiological responses to Bev in the training and validation sets, respectively. Responders (N = 105) and non-responders (N = 47) showed a 71.8% (IQR: 54.7-87.5) and 13.2% (IQR: 7.98-21.4) reduction in RN volumes, respectively. Five optimal clusters were determined using the shoulder method with ΔK metric. Spatial analyses revealed that 2 of the 5 clusters were associated with the edema areas, with 92.6% and 74.3% of the voxels located within this anatomic subregion. The two corresponding clusters were significantly associated with a response to Bev (Odds ratio [OR]: 11.12 (95% CI: 2.54-73.47), P = 0.004; OR: 1.63 [1.07-2.78], P = 0.042). Finally, we developed and validated a clinicoradiomics model based on features extracted from the 2 clusters that improved the prediction of Bev response, compared with a clinical-only model for both the training (AUC 0.852 vs 0.755) and validation (AUC 0.816 vs 0.691) cohorts. Conclusions: Our spatial radiomics approach revealed anatomic subregions that contain relevant radiomics features that could predict Bev efficacy in the treatment of RN in NPC patients. This method would improve the interpretability of radiomics, compared with current methods that lack spatial resolution.

Alzheimer's Disease Diagnosis and Biomarker Analysis Using Resting-State Functional MRI Functional Brain Network With Multi-Measures Features and Hippocampal Subfield and Amygdala Volume of Structural MRI.

Accurate diagnosis of the initial phase of Alzheimer’s disease (AD) is essential and crucial. The objective of this research was to employ efficient biomarkers for the diagnostic analysis and classification of AD based on combining structural MRI (sMRI) and resting-state functional MRI (rs-fMRI). So far, several anatomical MRI imaging markers for AD diagnosis have been identified. The use of cortical and subcortical volumes, the hippocampus, and amygdala volume, as well as genetic patterns, has proven to be beneficial in distinguishing patients with AD from the healthy population. The fMRI time series data have the potential for specific numerical information as well as dynamic temporal information. Voxel and graphical analyses have gained popularity for analyzing neurodegenerative diseases, such as Alzheimer’s and its prodromal phase, mild cognitive impairment (MCI). So far, these approaches have been utilized separately for the diagnosis of AD. In recent studies, the classification of cases of MCI into those that are not converted for a certain period as stable MCI (MCIs) and those that converted to AD as MCIc has been less commonly reported with inconsistent results. In this study, we verified and validated the potency of a proposed diagnostic framework to identify AD and differentiate MCIs from MCIc by utilizing the efficient biomarkers obtained from sMRI, along with functional brain networks of the frequency range .01–.027 at the resting state and the voxel-based features. The latter mainly included default mode networks (amplitude of low-frequency fluctuation [ALFF], fractional ALFF [ALFF], and regional homogeneity [ReHo]), degree centrality (DC), and salience networks (SN). Pearson’s correlation coefficient for measuring fMRI functional networks has proven to be an efficient means for disease diagnosis. We applied the graph theory to calculate nodal features (nodal degree [ND], nodal path length [NL], and between centrality [BC]) as a graphical feature and analyzed the connectivity link between different brain regions. We extracted three-dimensional (3D) patterns to calculate regional coherence and then implement a univariate statistical t-test to access a 3D mask that preserves voxels showing significant changes. Similarly, from sMRI, we calculated the hippocampal subfield and amygdala nuclei volume using Freesurfer (version 6). Finally, we implemented and compared the different feature selection algorithms to integrate the structural features, brain networks, and voxel features to optimize the diagnostic identifications of AD using support vector machine (SVM) classifiers. We also compared the performance of SVM with Random Forest (RF) classifiers. The obtained results demonstrated the potency of our framework, wherein a combination of the hippocampal subfield, the amygdala volume, and brain networks with multiple measures of rs-fMRI could significantly enhance the accuracy of other approaches in diagnosing AD. The accuracy obtained by the proposed method was reported for binary classification. More importantly, the classification results of the less commonly reported MCIs vs. MCIc improved significantly. However, this research involved only the AD Neuroimaging Initiative (ADNI) cohort to focus on the diagnosis of AD advancement by integrating sMRI and fMRI. Hence, the study’s primary disadvantage is its small sample size. In this case, the dataset we utilized did not fully reflect the whole population. As a result, we cannot guarantee that our findings will be applicable to other populations.

Research on Voxel-Based Features Detection and Analysis of Alzheimer’s Disease Using Random Survey Support Vector Machine

Alzheimer’s disease (AD) is a degenerative disease of the central nervous system characterized by memory and cognitive dysfunction, as well as abnormal changes in behavior and personality. The research focused on how machine learning classified AD became a recent hotspot. In this study, we proposed a novel voxel-based feature detection framework for AD. Specifically, using 649 voxel-based morphometry (VBM) methods obtained from MRI in Alzheimer’s Disease Neuroimaging Initiative (ADNI), we proposed a feature detection method according to the Random Survey Support Vector Machines (RS-SVM) and combined the research process based on image-, gene-, and pathway-level analysis for AD prediction. Particularly, we constructed 136, 141, and 113 novel voxel-based features for EMCI (early mild cognitive impairment)-HC (healthy control), LMCI (late mild cognitive impairment)-HC, and AD-HC groups, respectively. We applied linear regression model, least absolute shrinkage and selection operator (Lasso), partial least squares (PLS), SVM, and RS-SVM five methods to test and compare the accuracy of these features in these three groups. The prediction accuracy of the AD-HC group using the RS-SVM method was higher than 90%. In addition, we performed functional analysis of the features to explain the biological significance. The experimental results using five machine learning indicate that the identified features are effective for AD and HC classification, the RS-SVM framework has the best classification accuracy, and our strategy can identify important brain regions for AD.

Cube of Space Sampling for 3D Model Retrieval

Since the number of 3D models is rapidly increasing, extracting better feature descriptors to represent 3D models is very challenging for effective 3D model retrieval. There are some problems in existing 3D model representation approaches. For example, many of them focus on the direct extraction of features or transforming 3D models into 2D images for feature extraction, which cannot effectively represent 3D models. In this paper, we propose a novel 3D model feature representation method that is a kind of voxelization method. It is based on the space-based concept, namely CSS (Cube of Space Sampling). The CSS method uses cube space 3D model sampling to extract global and local features of 3D models. The experiments using the ESB dataset show that the proposed method to extract the voxel-based features can provide better classification accuracy than SVM and comparable retrieval results using the state-of-the-art 3D model feature representation method.

Targeted Rock Slope Assessment Using Voxels and Object-Oriented Classification

Reality capture technologies, also known as close-range sensing, have been increasingly popular within the field of engineering geology and particularly rock slope management. Such technologies provide accurate and high-resolution n-dimensional spatial representations of our physical world, known as 3D point clouds, that are mainly used for visualization and monitoring purposes. To extract knowledge from point clouds and inform decision-making within rock slope management systems, semantic injection through automated processes is necessary. In this paper, we propose a model that utilizes a segmentation procedure which delivers segments ready to classify and be retained or rejected according to complementary knowledge-based filter criteria. First, we provide relevant voxel-based features based on the local dimensionality, orientation, and topology and partition them in an assembly of homogenous segments. Subsequently, we build a decision tree that utilizes geometrical, topological, and contextual information and enables the classification of a multi-hazard railway rock slope section in British Columbia, Canada into classes involved in landslide risk management. Finally, the approach is compared to machine learning integrating recent featuring strategies for rock slope classification with limited training data (which is usually the case). This alternative to machine learning semantic segmentation approaches reduces substantially the model size and complexity and provides an adaptable framework for tailored decision-making systems leveraging rock slope semantics.

Rapid 3D phenotypic analysis of neurons and organoids using data-driven cell segmentation-free machine learning.

Phenotypic profiling of large three-dimensional microscopy data sets has not been widely adopted due to the challenges posed by cell segmentation and feature selection. The computational demands of automated processing further limit analysis of hard-to-segment images such as of neurons and organoids. Here we describe a comprehensive shallow-learning framework for automated quantitative phenotyping of three-dimensional (3D) image data using unsupervised data-driven voxel-based feature learning, which enables computationally facile classification, clustering and advanced data visualization. We demonstrate the analysis potential on complex 3D images by investigating the phenotypic alterations of: neurons in response to apoptosis-inducing treatments and morphogenesis for oncogene-expressing human mammary gland acinar organoids. Our novel implementation of image analysis algorithms called Phindr3D allowed rapid implementation of data-driven voxel-based feature learning into 3D high content analysis (HCA) operations and constitutes a major practical advance as the computed assignments represent the biology while preserving the heterogeneity of the underlying data. Phindr3D is provided as Matlab code and as a stand-alone program (https://github.com/DWALab/Phindr3D).

Convolutional neural networks for classification of Alzheimer's disease: Overview and reproducible evaluation.

Numerous machine learning (ML) approaches have been proposed for automatic classification of Alzheimer's disease (AD) from brain imaging data. In particular, over 30 papers have proposed to use convolutional neural networks (CNN) for AD classification from anatomical MRI. However, the classification performance is difficult to compare across studies due to variations in components such as participant selection, image preprocessing or validation procedure. Moreover, these studies are hardly reproducible because their frameworks are not publicly accessible and because implementation details are lacking. Lastly, some of these papers may report a biased performance due to inadequate or unclear validation or model selection procedures. In the present work, we aim to address these limitations through three main contributions. First, we performed a systematic literature review. We identified four main types of approaches: i) 2D slice-level, ii) 3D patch-level, iii) ROI-based and iv) 3D subject-level CNN. Moreover, we found that more than half of the surveyed papers may have suffered from data leakage and thus reported biased performance. Our second contribution is the extension of our open-source framework for classification of AD using CNN and T1-weighted MRI. The framework comprises previously developed tools to automatically convert ADNI, AIBL and OASIS data into the BIDS standard, and a modular set of image preprocessing procedures, classification architectures and evaluation procedures dedicated to deep learning. Finally, we used this framework to rigorously compare different CNN architectures. The data was split into training/validation/test sets at the very beginning and only the training/validation sets were used for model selection. To avoid any overfitting, the test sets were left untouched until the end of the peer-review process. Overall, the different 3D approaches (3D-subject, 3D-ROI, 3D-patch) achieved similar performances while that of the 2D slice approach was lower. Of note, the different CNN approaches did not perform better than a SVM with voxel-based features. The different approaches generalized well to similar populations but not to datasets with different inclusion criteria or demographical characteristics. All the code of the framework and the experiments is publicly available: general-purpose tools have been integrated into the Clinica software (www.clinica.run) and the paper-specific code is available at: https://github.com/aramis-lab/AD-DL.

Texture descriptors and voxels for the early diagnosis of Alzheimer's disease.

Early and accurate diagnosis of Alzheimer's Disease (AD) is critical since early treatment effectively slows the progression of the disease thereby adding productive years to those afflicted by this disease. A major problem encountered in the classification of MRI for the automatic diagnosis of AD is the so-called curse-of-dimensionality, which is a consequence of the high dimensionality of MRI feature vectors and the low number of training patterns available in most MRI datasets relevant to AD. A method for performing early diagnosis of AD is proposed that combines a set of SVMs trained on different texture descriptors (which reduce dimensionality) extracted from slices of Magnetic Resonance Image (MRI) with a set of SVMs trained on markers built from the voxels of MRIs. The dimension of the voxel-based features is reduced by using different feature selection algorithms, each of which trains a separate SVM. These two sets of SVMs are then combined by weighted-sum rule for a final decision. Experimental results show that 2D texture descriptors improve the performance of state-of-the-art voxel-based methods. The evaluation of our system on the four ADNI datasets demonstrates the efficacy of the proposed ensemble and demonstrates a contribution to the accurate prediction of AD. Ensembles of texture descriptors combine partially uncorrelated information with respect to standard approaches based on voxels, feature selection, and classification by SVM. In other words, the fusion of a system based on voxels and an ensemble of texture descriptors enhances the performance of voxel-based approaches.

Voxel-based 3D Point Cloud Semantic Segmentation: Unsupervised Geometric and Relationship Featuring vs Deep Learning Methods

Automation in point cloud data processing is central in knowledge discovery within decision-making systems. The definition of relevant features is often key for segmentation and classification, with automated workflows presenting the main challenges. In this paper, we propose a voxel-based feature engineering that better characterize point clusters and provide strong support to supervised or unsupervised classification. We provide different feature generalization levels to permit interoperable frameworks. First, we recommend a shape-based feature set (SF1) that only leverages the raw X, Y, Z attributes of any point cloud. Afterwards, we derive relationship and topology between voxel entities to obtain a three-dimensional (3D) structural connectivity feature set (SF2). Finally, we provide a knowledge-based decision tree to permit infrastructure-related classification. We study SF1/SF2 synergy on a new semantic segmentation framework for the constitution of a higher semantic representation of point clouds in relevant clusters. Finally, we benchmark the approach against novel and best-performing deep-learning methods while using the full S3DIS dataset. We highlight good performances, easy-integration, and high F1-score (&gt; 85%) for planar-dominant classes that are comparable to state-of-the-art deep learning.