Deep Learning-based Pipeline Research Articles

A Deep Learning-based Pipeline for Segmenting the Cerebral Cortex Laminar Structure in Histology Images.

Characterizing the anatomical structure and connectivity between cortical regions is a critical step towards understanding the information processing properties of the brain and will help provide insight into the nature of neurological disorders. A key feature of the mammalian cerebral cortex is its laminar structure. Identifying these layers in neuroimaging data is important for understanding their global structure and to help understand the connectivity patterns of neurons in the brain. We studied Nissl-stained and myelin-stained slice images of the brain of the common marmoset (Callithrix jacchus), which is a new world monkey that is becoming increasingly popular in the neuroscience community as an object of study. We present a novel computational framework that first acquired the cortical labels using AI-based tools followed by a trained deep learning model to segment cerebral cortical layers. We obtained a Euclidean distance of for the cortical labels acquisition, which was in the acceptable range by computing the half Euclidean distance of the average cortex thickness ( ). We compared our cortical layer segmentation pipeline with the pipeline proposed by Wagstyl et al. (PLoS biology, 18(4), e3000678 2020) adapted to 2D data. We obtained a better mean percentile Hausdorff distance (95HD) of . Whereas a mean 95HD of was obtained from Wagstyl et al. We also compared our pipeline's performance against theirs using their dataset (the BigBrain dataset). The results also showed better segmentation quality, Jaccard Index acquired from our pipeline, while was stated in their paper.

"Navigating the complexities of low-Grade glioma treatment: insights into SBT I-125 and novel assessment tools".

Central nervous system tumors, classified by the WHO into four grades based on their aggressiveness, present significant challenges in treatment, particularly low-grade gliomas (LGGs) which, despite their slower growth, can progress to high-grade gliomas. Lucca B. Palavani and colleagues evaluated the efficacy and safety of SBT I-125 brachytherapy for LGMs in a systematic review and meta-analysis of 20 studies involving 988 patients. The analysis revealed an overall complication rate of 10%, with headaches and cyst formation being the most frequent issues. The five-year progression-free survival (PFS) rate was 66%, while the ten-year PFS rate was 30%, and the rate of malignant transformation was 26%. The mortality rate was 33%. Despite these findings, significant limitations were noted, including data insufficiencies, study heterogeneity, lack of randomized controlled trials, and potential publication bias. Inconsistencies in follow-up durations further hindered the evaluation of long-term efficacy and safety. Recent advancements in automated tumor assessment, such as Cheng et al.'s deep learning-based pipeline, are revolutionizing glioma management by enhancing the accuracy and consistency of tumor volume and RANO measurements. These innovations facilitate improved glioma grading, genetic mutation prediction, surgical planning, real-time intraoperative guidance, and histopathological analysis. Integrating such advanced tools into clinical practice can significantly enhance the precision and efficiency of glioma management. In conclusion, while SBT I-125 brachytherapy shows promise, concerns regarding safety and efficacy underscore the need for further research with standardized methodologies. Incorporating advanced automated assessment tools could improve treatment evaluation and patient outcomes.

Accuracy and time efficiency of a novel deep learning algorithm for Intracranial Hemorrhage detection in CT Scans

PurposeTo evaluate a deep learning-based pipeline using a Dense-UNet architecture for the assessment of acute intracranial hemorrhage (ICH) on non-contrast computed tomography (NCCT) head scans after traumatic brain injury (TBI).Materials and methodsThis retrospective study was conducted using a prototype algorithm that evaluated 502 NCCT head scans with ICH in context of TBI.Four board-certified radiologists evaluated in consensus the CT scans to establish the standard of reference for hemorrhage presence and type of ICH. Consequently, all CT scans were independently analyzed by the algorithm and a board-certified radiologist to assess the presence and type of ICH. Additionally, the time to diagnosis was measured for both methods.ResultsA total of 405/502 patients presented ICH classified in the following types: intraparenchymal (n = 172); intraventricular (n = 26); subarachnoid (n = 163); subdural (n = 178); and epidural (n = 15). The algorithm showed high diagnostic accuracy (91.24%) for the assessment of ICH with a sensitivity of 90.37% and specificity of 94.85%. To distinguish the different ICH types, the algorithm had a sensitivity of 93.47% and a specificity of 99.79%, with an accuracy of 98.54%. To detect midline shift, the algorithm had a sensitivity of 100%. In terms of processing time, the algorithm was significantly faster compared to the radiologist’s time to first diagnosis (15.37 ± 1.85 vs 277 ± 14 s, p < 0.001).ConclusionA novel deep learning algorithm can provide high diagnostic accuracy for the identification and classification of ICH from unenhanced CT scans, combined with short processing times. This has the potential to assist and improve radiologists’ ICH assessment in NCCT scans, especially in emergency scenarios, when time efficiency is needed.

Real-Time Personal Protective Equipment Non-Compliance Recognition on AI Edge Cameras

Despite advancements in technology, safety equipment, and training within the construction industry over recent decades, the prevalence of fatal and nonfatal injuries and accidents remains a significant concern among construction workers. Hard hats and safety vests are crucial safety gear known to mitigate severe head trauma and other injuries. However, adherence to safety protocols, including the use of such gear, is often inadequate, posing potential risks to workers. Moreover, current manual safety monitoring systems are laborious and time-consuming. To address these challenges and enhance workplace safety, there is a pressing need to automate safety monitoring processes economically, with reduced processing times. This research proposes a deep learning-based pipeline for real-time identification of non-compliance with wearing hard hats and safety vests, enabling safety officers to preempt hazards and mitigate risks at construction sites. We evaluate various neural networks for edge deployment and find that the Single Shot Multibox Detector (SSD) MobileNet V2 model excels in computational efficiency, making it particularly suitable for this application-oriented task. The experiments and comparative analyses demonstrate the pipeline’s effectiveness in accurately identifying instances of non-compliance across different scenarios, underscoring its potential for improving safety outcomes.

Towards automated field ballast condition evaluation: Field validation of the ballast scanning vehicle capabilities

Ballast degradation can lead to adverse effects such as inadequate drainage, track settlement and reduced lateral stability, which could compromise track safety, daily functionality, and long-term maintenance. Field inspection of ballast for monitoring degradation and functional performance is a challenging task. Current state-of-the-practice methods for evaluating ballast primarily depend on subjective visual inspection, labor-intensive sampling, laboratory sieve analyses or Ground Penetrating Radar (GPR) technology. These methods fall short in providing an in-depth assessment of ballast, specifically in determining the degradation level and aggregate size and shape characteristics at various depths. In this regard, this research developed an innovative ballast investigation platform, the Ballast Scanning Vehicle (BSV), to automate the processes of acquiring detailed ballast inspection data. The BSV utilizes a deep learning-based pipeline for image segmentation to evaluate task-specific metrics such as coarse aggregate gradation, Fouling Index (FI), and continuous track FI depth profiles. This paper provides a detailed overview of the BSV’s functions as well as the different modules of the deep learning-based pipeline. Validation of the BSV’s capabilities was conducted at the Transportation Technology Center (TTC) and is discussed in detail. Based on the field results, the BSV is capable of providing accurate and near real-time evaluation of in-service ballast conditions, serves as a robust means for inspecting long sections of track, and can be used to investigate persistent trouble-spots related to track performance.

Experimental Verification on Deep Learning based Monitoring Algorithms for Early Detection of Damage in Buried Pipelines

Recent increases in buried pipeline damage accidents due to third-party interference have significantly heightened attention towards buried pipeline monitoring. Especially, as the sudden damage can lead to large-scale leakage, there is a necessity for preemptive response and maintenance. However, the application of a structural health monitoring approach is difficult, since the extensive network of buried pipelines, stretching over thousands of kilometers, exhibits diverse noise environments and propagation characteristics. As a result, challenges within the buried pipeline system frequently lead to damages being overlooked. In this study, introduces a deep learning-based pipeline damage monitoring algorithm, specifically designed to early detection of accidents caused by third-party interference. This algorithm integrates a CNN-based anomaly detection model, advanced signal processing for data preprocessing, and TDoA-based source localization. The training and test data set are the acquisition under completely independent conditions, which has been experimentally validate for applicability across various environments for buried pipelines. Moreover, both the training and test dataset acquisition were performed using accelerometers on in-service buried pipelines, each with diameters of 1,100 mm, 1,200 mm, and 2,200 mm, extending over lengths ranging from approximately 200 to 500 meters. Despite the independent conditions of the datasets, our study yielded over 95% accuracy in early detection, with the results being in good agreement with the actual excavate locations.

Segmentation and Multi-Timepoint Tracking of 3D Cancer Organoids from Optical Coherence Tomography Images Using Deep Neural Networks.

Recent years have ushered in a transformative era in in vitro modeling with the advent of organoids, three-dimensional structures derived from stem cells or patient tumor cells. Still, fully harnessing the potential of organoids requires advanced imaging technologies and analytical tools to quantitatively monitor organoid growth. Optical coherence tomography (OCT) is a promising imaging modality for organoid analysis due to its high-resolution, label-free, non-destructive, and real-time 3D imaging capabilities, but accurately identifying and quantifying organoids in OCT images remain challenging due to various factors. Here, we propose an automatic deep learning-based pipeline with convolutional neural networks that synergistically includes optimized preprocessing steps, the implementation of a state-of-the-art deep learning model, and ad-hoc postprocessing methods, showcasing good generalizability and tracking capabilities over an extended period of 13 days. The proposed tracking algorithm thoroughly documents organoid evolution, utilizing reference volumes, a dual branch analysis, key attribute evaluation, and probability scoring for match identification. The proposed comprehensive approach enables the accurate tracking of organoid growth and morphological changes over time, advancing organoid analysis and serving as a solid foundation for future studies for drug screening and tumor drug sensitivity detection based on organoids.

Enhancement of OCT en face images by unsupervised deep learning

Objective. The quality of optical coherence tomography (OCT) en face images is crucial for clinical visualization of early disease. As a three dimensional and coherent imaging, defocus and speckle noise are inevitable, which seriously affect evaluation of microstructure of bio-samples in OCT images. The deep learning has demonstrated great potential in OCT refocusing and denoising, but it is limited by the difficulty of sufficient paired training data. This work aims to develop an unsupervised method to enhance the quality of OCT en face images. Approach. We proposed an unsupervised deep learning-based pipeline. The unregistered defocused conventional OCT images and focused speckle-free OCT images were collected by a home-made speckle modulating OCT system to construct the dataset. The image enhancement model was trained with the cycle training strategy. Finally, the speckle noise and defocus were both effectively improved. Main results. The experimental results on complex bio-samples indicated that the proposed method is effective and generalized in enhancing the quality of OCT en face images. Significance. The proposed unsupervised deep learning method helps to reduce the complexity of data construction, which is conducive to practical applications in OCT bio-sample imaging.

A deep learning model for brain segmentation across pediatric and adult populations.

Automated quantification of brain tissues on MR images has greatly contributed to the diagnosis and follow-up of neurological pathologies across various life stages. However, existing solutions are specifically designed for certain age ranges, limiting their applicability in monitoring brain development from infancy to late adulthood. This retrospective study aims to develop and validate a brain segmentation model across pediatric and adult populations. First, we trained a deep learning model to segment tissues and brain structures using T1-weighted MR images from 390 patients (age range: 2-81 years) across four different datasets. Subsequently, the model was validated on a cohort of 280 patients from six distinct test datasets (age range: 4-90 years). In the initial experiment, the proposed deep learning-based pipeline, icobrain-dl, demonstrated segmentation accuracy comparable to both pediatric and adult-specific models across diverse age groups. Subsequently, we evaluated intra- and inter-scanner variability in measurements of various tissues and structures in both pediatric and adult populations computed by icobrain-dl. Results demonstrated significantly higher reproducibility compared to similar brain quantification tools, including childmetrix, FastSurfer, and the medical device icobrain v5.9 (p-value< 0.01). Finally, we explored the potential clinical applications of icobrain-dl measurements in diagnosing pediatric patients with Cerebral Visual Impairment and adult patients with Alzheimer's Disease.

Deep learning-based harmonization and super-resolution of Landsat-8 and Sentinel-2 images

Multi-spectral satellite images of the Earth’s surface are used in various applications, from water quality assessment and urban planning to climate monitoring, disaster response, infrastructure oversight, and agricultural surveillance. Many of these applications would benefit from higher spatial and temporal resolution of observations, which could be achieved by combining observations from several sources. This study introduces a deep learning-based pipeline to harmonize the spectral and spatial discrepancies between the Landsat-8 and Sentinel-2 Earth Observation satellites. Through established image quality metrics, we demonstrate a significant enhancement in the spatial resolution of Landsat-8 images. Field observation experiments show that leveraging unified images from both satellites increases the availability of cloud-free images by 21% annually on average in our study area. Additionally, our pipeline enhances the Normalized Difference Vegetation Index (NDVI) correlation between Landsat-8 and Sentinel-2 observations by about 4.9%, offering significant performance gains in a downstream crop segmentation task. Our 100M parameter model, trained on European data, generalizes to most regions with only minor limitations. Furthermore, we show that the pipeline can provide uncertainty estimates for its outputs, which are valuable for decision-making in downstream applications. Our code base is available online: https://github.com/venkatesh-thiru/Deep-Harmonization.

From data to knowledge: Construction process analysis through continuous image capturing, object detection, and knowledge graph creation

Persistent issues of schedule deviations and cost overruns within large construction projects aggravate the construction industry’s global productivity concerns. However, how holistic, data-oriented methods can effectively be leveraged for investigating project performance and identifying potential bottlenecks during the construction phase remains unanswered. Our research addresses this issue with a novel approach encompassing data acquisition, object detection, geometric projection, and graph-based linking. Image data, continuously captured by crane-camera systems, gets transformed into higher-level information using an end-to-end deep learning-based pipeline that covers the detection of specific on-site objects and integrates it in a knowledge graph. The knowledge graph facilitates extracting precise construction metrics, identifying spatiotemporal irregularities, like work hotspots characterized by high activity and intensive work concentrations, but also phases with low activity. The proposed method improves learning from past construction data, aiding stakeholders and inspiring further research into real-time monitoring, predictive analytics, and data-integrated decision-making systems to reshape construction practices.

A deep learning-based pipeline for developing multi-rib shape generative model with populational percentiles or anthropometrics as predictors

Rib cross-sectional shapes (characterized by the outer contour and cortical bone thickness) affect the rib mechanical response under impact loading, thereby influence the rib injury pattern and risk. A statistical description of the rib shapes or their correlations to anthropometrics is a prerequisite to the development of numerical human body models representing target demographics. Variational autoencoders (VAE) as anatomical shape generators remain to be explored in terms of utilizing the latent vectors to control or interpret the representativeness of the generated results. In this paper, we propose a pipeline for developing a multi-rib cross-sectional shape generative model from CT images, which consists of the achievement of rib cross-sectional shape data from CT images using an anatomical indexing system and regular grids, and a unified framework to fit shape distributions and associate shapes to anthropometrics for different rib categories. Specifically, we collected CT images including 3193 ribs, surface regular grid is generated for each rib based on anatomical coordinates, the rib cross-sectional shapes are characterized by nodal coordinates and cortical bone thickness. The tensor structure of shape data based on regular grids enable the implementation of CNNs in the conditional variational autoencoder (CVAE). The CVAE is trained against an auxiliary classifier to decouple the low-dimensional representations of the inter- and intra- variations and fit each intra-variation by a Gaussian distribution simultaneously. Random tree regressors are further leveraged to associate each continuous intra-class space with the corresponding anthropometrics of the subjects, i.e., age, height and weight. As a result, with the rib class labels and the latent vectors sampled from Gaussian distributions or predicted from anthropometrics as the inputs, the decoder can generate valid rib cross-sectional shapes of given class labels (male/female, 2nd to 11th ribs) for arbitrary populational percentiles or specific age, height and weight, which paves the road for future biomedical and biomechanical studies considering the diversity of rib shapes across the population.

Deep learning-based automatic pipeline for 3D needle localization on intra-procedural 3D MRI.

Accurate and rapid needle localization on 3D magnetic resonance imaging (MRI) is critical for MRI-guided percutaneous interventions. The current workflow requires manual needle localization on 3D MRI, which is time-consuming and cumbersome. Automatic methods using 2D deep learning networks for needle segmentation require manual image plane localization, while 3D networks are challenged by the need for sufficient training datasets. This work aimed to develop an automatic deep learning-based pipeline for accurate and rapid 3D needle localization on in vivo intra-procedural 3D MRI using a limited training dataset. The proposed automatic pipeline adopted Shifted Window (Swin) Transformers and employed a coarse-to-fine segmentation strategy: (1) initial 3D needle feature segmentation with 3D Swin UNEt TRansfomer (UNETR); (2) generation of a 2D reformatted image containing the needle feature; (3) fine 2D needle feature segmentation with 2D Swin Transformer and calculation of 3D needle tip position and axis orientation. Pre-training and data augmentation were performed to improve network training. The pipeline was evaluated via cross-validation with 49 in vivo intra-procedural 3D MR images from preclinical pig experiments. The needle tip and axis localization errors were compared with human intra-reader variation using the Wilcoxon signed rank test, with p < 0.05 considered significant. The average end-to-end computational time for the pipeline was 6s per 3D volume. The median Dice scores of the 3D Swin UNETR and 2D Swin Transformer in the pipeline were 0.80 and 0.93, respectively. The median 3D needle tip and axis localization errors were 1.48mm (1.09 pixels) and 0.98°, respectively. Needle tip localization errors were significantly smaller than human intra-reader variation (median 1.70mm; p < 0.01). The proposed automatic pipeline achieved rapid pixel-level 3D needle localization on intra-procedural 3D MRI without requiring a large 3D training dataset and has the potential to assist MRI-guided percutaneous interventions.

Abstract 3536: Paracell: A high throughput, deep learning-based pipeline for single-cell phenotypic profiling

Abstract Variants of uncertain significance constitute the vast majority of missense variants, hampering the utility of genetic testing, despite its remarkable impacts on clinical diagnosis and decision-making. Although the sequencing and interpretation of genetic variants have been drastically improved in recent years, only about 2% of novel missense variants are currently clinically actionable. Therefore, a robust and scalable pipeline is desperately needed to provide functional evidence critical for variant interpretation. In this work, we demonstrate Paracell, a deep learning-based phenotypic profiling pipeline that leverages subcellular segmentation, high-dimensional single-cell phenotyping, and machine learning to classify the functional impact of variants. Paracell captures heterogeneity in cellular phenotypes and protein colocalization between variants and their signaling partners, as well as subcellular markers, and enables analysis at a single-cell resolution. Using a classification model trained on single-cell features extracted by Paracell, we were able to accurately classify cells from loss-of-function (LoF) variants based on their impact. After aggregating cellular profiles on a variant level, our pipeline was able to distinguish LoF variants from functional ones with high sensitivity and specificity. Our work demonstrates that Paracell is a robust and scalable method that can sensitively detect differences in single-cell phenotypic profiles. The systematic application of this pipeline will provide valuable functional evidence for variant interpretation, enhancing their clinical utility and accelerating personalized cancer care. Citation Format: David L. Nguyen, Jesse T. Chao. Paracell: A high throughput, deep learning-based pipeline for single-cell phenotypic profiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3536.

Abstract 5488: AI-enhanced spatial proteomics reveals B-cells &amp; FoxP3+ cells interactions in HPV(+) oropharyngeal squamous cell carcinoma Progression

Abstract Background: Analyzing interactions and the microanatomical distribution of the tumor immune microenvironment (TIME) is essential for a comprehensive understanding of tumor progression. Imaging Mass Cytometry (IMC) is a high-dimensional tissue imaging system that enables the deeper and multiparametric in situ exploration of tumor microenvironments at a single-cell level. This study describes a pipeline for analyzing 39-plex IMC images of human HPV(+) oropharyngeal squamous cell carcinoma (OPSCC) tumor sections, elucidating immune cells' nature, functions, and interactions with tumor cells. The IMC data analysis provides valuable information for clinical studies that could be used for the identification of prognostic biomarkers and mechanisms of resistance to current immunotherapies. Methods: A retrospective study was conducted on (N=20) patients (10 progressors and 10 non-progressors) with HPV(+) OPSCC. The dataset comprises paired hematoxylin and eosin (H&amp;E) images with IMC data from multiple Regions of Interest (ROIs) (N=82), enabling the in-situ profiling of 39 proteins in primary tumor tissue. Cell segmentation was followed by employing original images and their segmentation masks for cell phenotype identification using a probabilistic neural network method to spatially profile cancer cells, macrophages, NK-M cells, B-cells, and distinct T-cell populations. We identified protein expression at single-cell resolution and investigated the impact of multicellular spatial organization on the disease progression. The spatial neighborhood interaction score of tumor cells with various other cells was computed for all the ROIs across the two groups. Results: The markers showed varying expression levels of intensity distribution, likely due to differences in tissue sample composition. Specifically, CD3, CD4, CD8, CD19, CD20, and CD68 showed significantly higher intensity in non-progressors. Interestingly, FoxP3 also had a significantly higher intensity level in non-progressors. Cell phenotyping emphasized that non-progressors generally have increased levels of NK-M cell markers. Spatial analysis revealed a significant increase in interaction between CD3 cells and tumor cells in non-progressors (p-value=0.07, Mann-Whitney U test). Furthermore, a significant rise in the interaction between B-cells and FoxP3+ cells was observed in non-progressors (p-value&amp;lt;0.05, Mann-Whitney U test) suggesting a coordinated suppression of tumor growth. Conclusion: We developed a deep learning-based pipeline to analyze spatial relationships in the TIME of IMC data, demonstrating its clinical relevance in predicting disease progression. Our findings emphasize the importance of spatial relationships in the TME for response and suggest that proximity between either CD3+ T-cells to cancer cells or B-cells to FoxP3+ cells are candidate biomarkers. Citation Format: Sumanth Reddy Nakkireddy, Routman David, Kathleen Bartemes, Daniel Ma, Tae Hyun Hwang, Kathryn Van Abel, Chadi Nimeh Abdel-Halim. AI-enhanced spatial proteomics reveals B-cells &amp; FoxP3+ cells interactions in HPV(+) oropharyngeal squamous cell carcinoma Progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5488.

A practical guide to the development and deployment of deep learning models for the orthopaedic surgeon: Part III, focus on registry creation, diagnosis,and data privacy.

Deep learning is a subset of artificial intelligence (AI) with enormous potential to transform orthopaedic surgery. As has already become evident with the deployment of Large Language Models (LLMs) like ChatGPT (OpenAI Inc.), deep learning can rapidly enter clinical and surgical practices. As such, it is imperative that orthopaedic surgeons acquire a deeper understanding of the technical terminology, capabilitiesand limitations associated with deep learning models. The focus of this series thus far has been providing surgeons with an overview of the steps needed to implement a deep learning-based pipeline, emphasizing some of the important technical details for surgeons to understand as they encounter, evaluateor lead deep learning projects. However, this series would be remiss without providing practical examples of how deep learning models have begun to be deployed and highlighting the areas where the authors feel deep learning may have the most profound potential. While computer vision applications of deep learning were the focus of Parts I and II, due to the enormous impact that natural language processing (NLP) has had in recent months, NLP-based deep learning models are also discussed in this final part of the series. In this review, three applications that the authors believe can be impacted the most by deep learning but with which many surgeons may not be familiar are discussed: (1) registry construction, (2) diagnostic AIand (3) data privacy. Deep learning-based registry construction will be essential for the development of more impactful clinical applications, with diagnostic AI being one of those applications likely to augment clinical decision-making in the near future. As the applications of deep learning continue to grow, the protection of patient information will become increasingly essential; as such, applications of deep learning to enhance data privacy are likely to become more important than ever before. Level of Evidence: Level IV.

Abstract TP143: Longitudinal Evaluation of Vessel Wall MRI Demonstrates Bilaterally Symmetric Evolution of Carotid Atherosclerosis

Introduction: Carotid atherosclerosis is a leading cause of stroke worldwide. Previous studies demonstrated bilateral symmetry in atherosclerotic plaque burden and calcification scores. By leveraging longitudinal 3D vessel wall MRI, this study aims to investigate if there is a symmetrical trend of bilateral plaque progression. Methods: Asymptomatic patients with confirmed carotid stenosis were prospectively recruited in the study. Subjects underwent baseline and follow-up 3D vessel wall imaging for a time span up to 8 years. A deep learning-based pipeline was used to bilaterally segment the carotid lumen and vessel wall on axial image slices. Within each artery, the range of slices with wall thickness &gt; 2mm on any scan was included as potential atherosclerotic plaque. Plaque burden, positively associated with stroke risk, was defined as the percent wall volume (wall volume / total vessel volume х 100%). For each plaque, a linear model was fitted across all available time points to determine the growth rate of plaque burden. We used generalized estimating equation-based linear regression models to evaluate associations of the growth rate of each side with the contralateral side, adjusted for clinical risk factors and medications. Effect sizes were summarized using the partial correlation coefficient. Results: 20 subjects (mean age 73.5 years, 3 females, 14 on statins) who underwent at least three scans with bilateral carotid plaque were included in the analysis. Each subject underwent an average of 4.8 scans over 5.2 years. Overall, the plaque growth rate of one side was moderately correlated with the contralateral side (r=0.50, p=0.004), even after adjusting for age, sex and statins use. Then, for each subject, the larger of the two plaques at baseline (left or right side) was identified. The growth rate of the larger plaque was also predictive of the growth rate of the smaller plaque over the same time period (r=0.66, p=0.004). Conclusion: The systemic nature of atherosclerotic disease entails an association in symmetrical disease evolution for left and right carotid arteries. In bilaterally diseased patients, the longitudinal monitoring of the more advanced carotid lesion may provide prognostic information on the less advanced contralateral plaque.

Deep Learning Algorithm of the SPARCC Scoring System in SI Joint MRI.

The Spondyloarthritis Research Consortium of Canada (SPARCC) scoring system is a sacroiliitis grading system. To develop a deep learning-based pipeline for grading sacroiliitis using the SPARCC scoring system. Prospective. The study included 389 participants (42.2-year-old, 44.6% female, 317/35/37 for training/validation/testing). A pretrained algorithm was used to differentiate image with/without sacroiliitis. 3-T, short tau inversion recovery (STIR) sequence, fast spine echo. The regions of interest as ground truth for models' training were identified by a rheumatologist (HYC, 10-year-experience) and a radiologist (KHL, 6-year-experience) using the Assessment of Spondyloarthritis International Society definition of MRI sacroiliitis independently. Another radiologist (YYL, 4.5-year-experience) solved the discrepancies. The bone marrow edema (BME) and sacroiliac region models were for segmentation. Frangi-filter detected vessels used as intense reference. Deep learning pipeline scored using SPARCC scoring system evaluating presence and features of BMEs. A rheumatologist (SCWC, 6-year-experience) and a radiologist (VWHL, 14-year-experience) scored using the SPARCC scoring system once. The radiologist (YYL) scored twice with 5-day interval. Independent samples t-tests and Chi-squared tests were used. Interobserver and intraobserver reliability by intraclass correlation coefficient (ICC) and Pearson coefficient evaluated consistency between readers and the deep learning pipeline. We evaluated the performance using sensitivity, accuracy, positive predictive value, and Dice coefficient. A P-value <0.05 was considered statistically significant. The ICC and the Pearson coefficient between the SPARCC scores from three readers and the deep learning pipeline were 0.83 and 0.86, respectively. The sensitivity in identifying BME and accuracy of identifying SI joints and blood vessels was 0.83, 0.90, and 0.88, respectively. The dice coefficients were 0.82 (sacrum) and 0.80 (ilium). The high consistency with human readers indicated that deep learning pipeline may provide a SPARCC-informed deep learning approach for scoring of STIR images in spondyloarthritis. 1 TECHNICAL EFFICACY: Stage 2.

Deep Learning and fMRI-Based Pipeline for Optimization of Deep Brain Stimulation During Parkinson's Disease Treatment: Toward Rapid Semi-Automated Stimulation Optimization.

Optimized deep brain stimulation (DBS) is fast becoming a therapy of choice for the treatment of Parkinson's disease (PD). However, the post-operative optimization (aimed at maximizing patient clinical benefits and minimizing adverse effects) of all possible DBS parameter settings using the standard-of-care clinical protocol requires numerous clinical visits, which substantially increases the time to optimization per patient (TPP), patient cost burden and limit the number of patients who can undergo DBS treatment. The TPP is further elongated in electrodes with stimulation directionality or in diseases with latency in clinical feedback. In this work, we proposed a deep learning and fMRI-based pipeline for DBS optimization that can potentially reduce the TPP from ~1 year to a few hours during a single clinical visit. We developed an unsupervised autoencoder (AE)-based model to extract meaningful features from 122 previously acquired blood oxygenated level dependent (BOLD) fMRI datasets from 39 a priori clinically optimized PD patients undergoing DBS therapy. The extracted features are then fed into multilayer perceptron (MLP)-based parameter classification and prediction models for rapid DBS parameter optimization. The AE-extracted features of optimal and non-optimal DBS were disentangled. The AE-MLP classification model yielded accuracy, precision, recall, F1 score, and combined AUC of 0.96 ± 0.04, 0.95 ± 0.07, 0.92 ± 0.07, 0.93 ± 0.06, and 0.98 respectively. Accuracies of 0.79 ± 0.04, 0.85 ± 0.04, 0.82 ± 0.05, 0.83 ± 0.05, and 0.70 ± 0.07 were obtained in the prediction of voltage, frequency, and x-y-z contact locations, respectively. The proposed AE-MLP models yielded promising results for fMRI-based DBS parameter classification and prediction, potentially facilitating rapid semi-automated DBS parameter optimization. Clinical and Translational Impact Statement-A deep learning-based pipeline for semi-automated DBS parameter optimization is presented, with the potential to significantly decrease the optimization duration per patient and patients' financial burden while increasing patient throughput.

HandKey: Knocking-Triggered Robust Vibration Signature for Keyless Unlocking

Door lock is regarded as a critical line of defending the privacy and security of personal areas. However, for inner doors in environments like factories, existing locking mechanisms can be poor in user-friendliness and high in cost. For instance, mechanical locks require carrying keys that inevitably compromise user experiences, while smart locks always require non-trivial sensors. Therefore, inner doors urgently require a lightweight unlocking scheme that can properly balance user-friendliness, cost, and security. To this end, we propose HandKey as a keyless unlocking scheme to supplement existing lock systems. HandKey relies on two principles: the simplicity of hand knocking doors and the uniqueness of vibration triggered by the knocking force. In other words, a door and a hand knocking it jointly form a unique physical system that generates hand-dependent and user-specific vibration signatures uniquely representing a user identity. In designing HandKey, we first analyze the vibration mechanism behind it and the impacts of gestures and door materials on vibration signatures. Then we innovatively construct a signal processing and deep learning-based pipeline to extract signatures robust to variable knocking behaviors for representing user identity. Finally, we implement a HandKey prototype and use extensive evaluation to demonstrate its security and effectiveness.