Deep Learning-based Segmentation Algorithms Research Articles

Development and validation of a deep learning-based method for automatic measurement of uterus, fibroid, and ablated volume in MRI after MR-HIFU treatment of uterine fibroids

IntroductionThe non-perfused volume divided by total fibroid load (NPV/TFL) is a predictive outcome parameter for MRI-guided high-intensity focused ultrasound (MR-HIFU) treatments of uterine fibroids, which is related to long-term symptom relief. In current clinical practice, the MR-HIFU outcome parameters are typically determined by visual inspection, so an automated computer-aided method could facilitate objective outcome quantification. The objective of this study was to develop and evaluate a deep learning-based segmentation algorithm for volume measurements of the uterus, uterine fibroids, and NPVs in MRI in order to automatically quantify the NPV/TFL. Materials and MethodsA segmentation pipeline was developed and evaluated using expert manual segmentations of MRI scans of 115 uterine fibroid patients, screened for and/or undergoing MR-HIFU treatment. The pipeline contained three separate neural networks, one per target structure. The first step in the pipeline was uterus segmentation from contrast-enhanced (CE)-T1w scans. This segmentation was subsequently used to remove non-uterus background tissue for NPV and fibroid segmentation. In the following step, NPVs were segmented from uterus-only CE-T1w scans. Finally, fibroids were segmented from uterus-only T2w scans. The segmentations were used to calculate the volume for each structure. Reliability and agreement between manual and automatic segmentations, volumes, and NPV/TFLs were assessed. ResultsFor treatment scans, the Dice similarity coefficients (DSC) between the manually and automatically obtained segmentations were 0.90 (uterus), 0.84 (NPV) and 0.74 (fibroid). Intraclass correlation coefficients (ICC) were 1.00 [0.99, 1.00] (uterus), 0.99 [0.98, 1.00] (NPV) and 0.98 [0.95, 0.99] (fibroid) between manually and automatically derived volumes. For manually and automatically derived NPV/TFLs, the mean difference was 5% [-41%, 51%] (ICC: 0.66 [0.32, 0.85]). ConclusionThe algorithm presented in this study automatically calculates uterus volume, fibroid load, and NPVs, which could lead to more objective outcome quantification after MR-HIFU treatments of uterine fibroids in comparison to visual inspection. When robustness has been ascertained in a future study, this tool may eventually be employed in clinical practice to automatically measure the NPV/TFL after MR-HIFU procedures of uterine fibroids.

MSA-UNet: A Multiscale Lightweight U-Net Lung CT Image Segmentation Algorithm Under Attention Mechanism

Automatic and precise segmentation of lung images can assist doctors in locating and diagnosing lung lesions. However, current traditional lung CT image lesion segmentation algorithms suffer from the problem of low segmentation accuracy, while deep learning-based segmentation algorithms struggle to strike a better balance between lightweight and high accuracy. In response to this issue, a multi-scale and lightweight U-Net lung image segmentation algorithm with an attention mechanism is proposed. This algorithm introduces CA convolution after the convolution in the encoding stage to extract channel relationships and positional information from the feature maps. Furthermore, the RFB module is employed to extract features from different perspectives. Lastly, upward residual connections are introduced between the RFB modules in the encoder and decoder to enhance inter-network information interaction. Experiments conducted on the LUNA (lung nodule analysis) dataset and the COVID-QU-Ex dataset for COVID-19 pneumonia demonstrate that the proposed MSA-UNet algorithm achieves the best results in terms of Precision and Dice metrics. It outperforms mainstream models such as U-Net++ and DeeplabV3+ in terms of segmentation effectiveness and segmentation generality. The model has a floating-point operation count (FLOPs) of 18.15 G, a network parameter counts of 8.83×106, and achieves a Precision of 99.37%. The algorithm achieves a good balance between computational efficiency, model size, and segmentation accuracy.

Federated Learning: A Cross-Institutional Feasibility Study of Deep Learning Based Intracranial Tumor Delineation Framework for Stereotactic Radiosurgery.

Deep learning-based segmentation algorithms usually required large or multi-institute data sets to improve the performance and ability of generalization. However, protecting patient privacy is a key concern in the multi-institutional studies when conventional centralized learning (CL) is used. To explores the feasibility of a proposed lesion delineation for stereotactic radiosurgery (SRS) scheme for federated learning (FL), which can solve decentralization and privacy protection concerns. Retrospective. 506 and 118 vestibular schwannoma patients aged 15-88 and 22-85 from two institutes, respectively; 1069 and 256 meningioma patients aged 12-91 and 23-85, respectively; 574 and 705 brain metastasis patients aged 26-92 and 28-89, respectively. 1.5T, spin-echo, and gradient-echo [Correction added after first online publication on 21 August 2023. Field Strength has been changed to "1.5T" from "5T" in this sentence.]. The proposed lesion delineation method was integrated into an FL framework, and CL models were established as the baseline. The effect of image standardization strategies was also explored. The dice coefficient was used to evaluate the segmentation between the predicted delineation and the ground truth, which was manual delineated by neurosurgeons and a neuroradiologist. The paired t-test was applied to compare the mean for the evaluated dice scores (p < 0.05). FL performed the comparable mean dice coefficient to CL for the testing set ofTaipei Veterans General Hospital regardless of standardization and parameter; for the Taichung Veterans General Hospital data, CL significantly (p < 0.05) outperformed FL while using bi-parameter, but comparable results while using single-parameter. For the non-SRS data, FL achieved the comparable applicability to CL with mean dice 0.78 versus 0.78 (without standardization), and outperformed to the baseline models of two institutes. The proposed lesion delineation successfully implemented into an FL framework. The FL models were applicable on SRS data of each participating institute, and the FL exhibited comparable mean dice coefficient to CL on non-SRS dataset. Standardization strategies would be recommended when FL is used. 4 TECHNICAL EFFICACY: Stage 1.

Quantification of geogrid lateral restraint using transparent sand and deep learning-based image segmentation

An experimental technique is presented to quantify the lateral restraint provided by a geogrid embedded in granular soil at the particle level. Repeated load triaxial tests were done on transparent sand specimens stiffened with geosynthetic inclusions simulating geogrids. Images of laser illuminated planes through the specimens were segmented into particles using a deep learning-based segmentation algorithm. The particle outlines were characterized in terms of Fourier shape descriptors and tracked across sequentially captured images. The accuracy of the particle displacement measurements was validated against Digital Image Correlation (DIC) measurements. In addition, the method's resolution and repeatability is presented. Based on the measured particle displacements and rotations, a state boundary line between probable and improbable particle motions was identified for each test. The size of the zone of probable motions was used to quantify the lateral restraint provided by the inclusions. Overall, the tests results revealed that the geosynthetic inclusions restricted both particle displacements and rotations. However, the particle displacements were found to be restrained more significantly than the rotations. Finally, a unique relationship was found between the magnitude of the permanent strains of the specimens and the area of the zone of probable motions.

Self-training adversarial learning for cross-domain retinal OCT fluid segmentation

Accurate measurements of the size, shape and volume of macular edema can provide important biomarkers to jointly assess disease progression and treatment outcome. Although many deep learning-based segmentation algorithms have achieved remarkable success in semantic segmentation, these methods have difficulty obtaining satisfactory segmentation results in retinal optical coherence tomography (OCT) fluid segmentation tasks due to low contrast, blurred boundaries, and varied distribution. Moreover, directly applying a well-trained model on one device to test the images from other devices may cause the performance degradation in the joint analysis of multi-domain OCT images. In this paper, we propose a self-training adversarial learning framework for unsupervised domain adaptation in retinal OCT fluid segmentation tasks. Specifically, we develop an image style transfer module and a fine-grained feature transfer module to reduce discrepancies in the appearance and high-level features of images from different devices. Importantly, we transfer the target images to the appearance of source images to ensure that no image information of the source domain for supervised training is lost. To capture specific features of the target domain, we design a self-training module based on a discrepancy and similarity strategy to select the images with better segmentation results from the target domain and then introduce them into the source domain for the iterative training segmentation model. Extensive experiments on two challenging datasets demonstrate the effectiveness of our proposed method. In Particular, our proposed method achieves comparable results on cross-domain retinal OCT fluid segmentation compared with the state-of-the-art methods.

Ship hull number detection and recognition under sparse samples

目的 舰船舷号检测识别是海面态势感知的关键技术，精准的舷号检测识别对海洋权益保护具有重要意义。但目前没有公开数据提供支持。为此，本文先构建了一个真实场景下的稀疏舰船舷号数据集（sparse ship hullnumber dataset in real scene，SSHN-RS），包含3 004幅舰船图像，共计11 328个舷号字符，覆盖了多国、各类、水平、倾斜、背景简单、背景复杂、光线不佳和被遮挡的舰船舷号样本，是一个具有挑战性的数据集。基于SSHN-RS，开展舰船舷号检测识别研究，其主要难点在于：1）样本稀疏，模型容易过拟合；2）舷号字符分布密集，网络难以充分提取各字符特征；3）部分字符存在嵌套区域和相似区域，网络会识别出大量冗余结果。针对上述难点，提出了一种基于多视角渐进式上下文解耦的舰船舷号检测识别算法。方法 首先，引入一个固定中心和最大化面积的随机透视变换技术，在不增加样本数量的前提下扩充舷号姿态，实现了数据增广，提升了模型的泛化能力；其次，提出了一个渐进式上下文解耦技术，先通过依次擦除舷号各字符生成一系列新样本，再利用特征提取网络提取和融合各样本的多尺度特征，不仅减少字符上下文信息对特征学习的干扰，而且再次增广了数据；最后，在测试阶段，提出了一个掩码间扰动抑制技术，先根据预测结果采用与渐进式上下文解耦技术类似的方法生成新样本并重新进行预测，再引入一个1维非极大值抑制技术去除预测结果中错误的冗余字符，输出最佳检测识别结果，进一步优化网络性能。结果 在SSHN-RS上采用主流实例分割算法进行定性和定量评估。在定量评估上，本文算法舷号的检测精确率、召回率、F值和识别率分别可达0.985 4，0.957 6，0.971 3，0.901 8，均优于其他算法。相比指标排名第2的算法，分别提高了4.51%，3.45%，3.97%，8.83%；在定性评估上，本文算法更适合舰船舷号检测识别任务，检测识别性能更高。此外，本文算法可以泛化到其他实例分割算法中，以经典算法Mask RCNN（mask region based convolutional neural network）为例，加入本文算法各模块后，各指标分别提升了9.82%，6.04%，7.80%，6.73%。结论 本文算法可以解决舷号检测识别任务中因样本稀疏、舷号分布密集、部分字符存在嵌套和相似性带来的问题，在主观和客观上均取得了最先进的性能，并且具有通用性。SSHN-RS可通过https://github.com/Bingchuan897/SSHN-RS获取。;Objective Ship hull number detection and recognition can be as the key technologies for marine awareness. It is essential for the preservation of maritime rights and interests. However，it is required to data-driven researches in support of ship hull number detection and recognition. Therefore，we develop a sparse ship hull number dataset in real scene （SSHN-RS），which contains 3 004 images with a total of 11 328 hull numbers. The challenging SSHN-RS dataset is featured of ship hull numbers of various countries，hull numbers of various types，horizontal hull numbers，inclined hull numbers，hull numbers with complex background，hull numbers with simple background，poorly illuminated hull numbers and partially occluded hull numbers. We carry out SSHN-RS-related research on ship hull number detection and recognition. The main challenges are required to be resolved on three aspects as following：1）the hull number samples are sparse， which causes over-fitting of the network，2）the features of hull number are densely distributed，which is challenged to learn some of the hull number characteristics fully，and 3）some hull number have its nested areas and a high degree of similarity，which is costly for large number of redundant results. Method To resolve these problems mentioned above，we demonstrate a ship hull number detection and recognition algorithm in terms of multi-view progressive context decoupling. First，a random perspective transformation technology with fixed center and maximized area is illustrated. To realize data augmentation and improve the generalization ability of the model，the hull number spatial attitude is extended without increasing the number of samples. Second，a progressive context decoupling technology is proposed. A series of new samples are first generated by sequentially erasing each character of the hull number，and the feature extraction network is then used to extract and fuse the multi-scale features of each sample. It can reduce the influence of feature-contextual information on feature learning and rich the data expansion. It can improve the feature expression ability effectively as well. Finally，in the testing stage，to generate new samples，and inputs the new samples into the testing network for prediction， an inter mask disturbance suppression technology is first focused on a method similar to the progressive context decoupling technology. At the same time，a one-dimensional non-maximum suppression technology is introduced that it can mainly process the hull number recognition results of each sample twice. To suppress the redundant mask in the detection and recognition results effectively，it can make each character of the hull number of all samples correspond to a recognition result. For the testing network，the output has only a set of optimal results. However，there is hull number noisy recognition in some samples. Therefore，the hull number recognition results of all samples are added on the original images，and the second one-dimensional non-maximum suppression technology is performed and processed on it. It can suppress the noises in some samples and outputs a set of optimal results. The post-processing module can optimize the detection and recognition performance further. Result The comparative experiment is mainly carried out on SSHN-RS. First，we conduct evaluation analysis on the general instance segmentation algorithms. Multi-view progressive context decoupling-based detection precision，recall，f-score and recognition rate of the ship number detection and recognition algorithm can reach 0. 985 4，0. 957 6， 0. 971 3 and 0. 901 8，which are improved by 4. 51%，3. 45%，3. 97% and 8. 83% respectively compared to the secondranked method. Second，the ablation experiments on the ship hull number detection and recognition algorithm are carried out in terms of multi-view progressive context decoupling. The experimental results show that the performance of ship hull number detection and recognition is improved whether each technology is used solely or mixed. Finally，to validate its effectiveness，we apply some popular modules of the ship hull number detection and recognition algorithm based on multiview progressive context decoupling to other general instance segmentation algorithms. Taking the classic algorithm mask region based convolutional neural network（Mask RCNN）as an example，the indexes have been improved by 9. 82%， 6. 04%，7. 80% and 6. 73% of each after our algorithm module is added. Conclusion Our SSHN-RS contains rich and effective ship hull number information，which can provide data support for ship hull number detection and recognition. The experimental results show that the multi-view progressive context decoupling-based ship hull number detection and recognition algorithm can optimize the detection and recognition performance of hull numbers due to sparse samples，dense character distribution，nested areas，and characters-between high similarity. This algorithm can be generalized to other deep learning based segmentation algorithms. The benchmarks-relevant are provided as a basis for future research on ship hull number detection and recognition. The dataset is available at https://github.com/Bingchuan897/SSHN-RS.

Deep Learning Techniques for Early Detection of Alzheimer’s Disease: A Review

Alzheimer's disease (AD) is the most prevalent kind of dementia illness that can significantly impair a person's capability to carry out everyday tasks. According to findings, AD may be the third provoking reason of mortality among older adults, behind cancer and heart disease. Individuals at risk of acquiring AD must be identified before treatment strategies may be tested. The study's goal is to give a thorough examination of tissue structures using segmented MRI, which will lead to a more accurately labeling of certain brain illnesses. Several complicated segmentation approaches for identify AD have been developed. DL algorithms for brain structure segmentation and AD categorization have gotten a lot of attention since they can deliver accurate findings over a huge amount of data. As a result, DL approaches are increasingly favored over cutting-edge Machine Learning (ML) techniques. This study provides you with an overview of current trend deep learning-based segmentation algorithms for analyzing brain Magnetic Resonance Imaging for the treatment of AD. Finally, a conversation on the approaches' benefits and drawbacks, as well as future directives, was held, which may help researchers better comprehend present algorithms and methods in this field, and eventually design new and more successful algorithms.

E-MPSPNet: Ice–Water SAR Scene Segmentation Based on Multi-Scale Semantic Features and Edge Supervision

Distinguishing sea ice and water is crucial for safe navigation and carrying out offshore activities in ice zones. However, due to the complexity and dynamics of the ice–water boundary, it is difficult for many deep learning-based segmentation algorithms to achieve accurate ice–water segmentation in synthetic aperture radar (SAR) images. In this paper, we propose an ice–water SAR segmentation network, E-MPSPNet, which can provide effective ice–water segmentation by fusing semantic features and edge information. The E-MPSPNet introduces a multi-scale attention mechanism to better fuse the ice–water semantic features and designs an edge supervision module (ESM) to learn ice–water edge features. The ESM not only provides ice–water edge prediction but also imposes constraints on the semantic feature extraction to better express the edge information. We also design a loss function that focuses on both ice–water edges and semantic segmentations of ice and water for overall network optimization. With the AI4Arctic/ASIP Sea Ice Dataset as the benchmark, experimental results show our E-MPSPNet achieves the best performance compared with other commonly used segmentation models, reaching 94.2% for accuracy, 93.0% for F-score, and 89.2% for MIoU. Moreover, our E-MPSPNet shows a relatively smaller model size and faster processing speed. The application of the E-MPSPNet for processing a SAR scene demonstrates its potential for operational use in drawing near real-time navigation charts of sea ice.

Evaluation of oocyte maturity using artificial intelligence quantification of follicle volume biomarker by three-dimensional ultrasound

Research questionCan a novel deep learning-based follicle volume biomarker using three-dimensional ultrasound (3D-US) be established to aid in the assessment of oocyte maturity, timing of HCG administration and the individual prediction of ovarian hyper-response? DesignA total of 515 IVF cases were enrolled, and 3D-US scanning was carried out on HCG administration day. A follicle volume biomarker established by means of a deep learning-based segmentation algorithm was used to calculate optimal leading follicle volume for predicting number of mature oocytes retrieved and optimizing HCG trigger timing. Performance of the novel biomarker cut-off value was compared with conventional two-dimensional ultrasound (2D-US) follicular diameter measurements in assessing oocyte retrieval outcome. Moreover, demographics, infertility work-up and ultrasound biomarkers were used to build models for predicting ovarian hyper-response. ResultsOn the basis of the deep learning method, the optimal cut-off value of the follicle volume biomarker was determined to be 0.5 cm3 for predicting number of mature oocytes retrieved; its performance was significantly better than the conventional method (two-dimensional diameter measurement ≥10 mm). The cut-off value for leading follicle volume to optimize HCG trigger timing was determined to be 3.0 cm3 and was significantly associated with a higher number of mature oocytes retrieved (P = 0.01). Accuracy of the multi-layer perceptron model was better than two-dimensional diameter measurement (0.890 versus 0.785) and other multivariate classifiers in predicting ovarian hyper-response (P < 0.001). ConclusionsDeep learning segmentation methods and multivariate classifiers based on 3D-US were found to be potentially effective approaches for assessing mature oocyte retrieval outcome and individual prediction of ovarian hyper-response.

Cross-sectional and longitudinal Biomarker extraction and analysis for multicentre FLAIR brain MRI

Fluid-attenuated inversion recovery (FLAIR) MRI has emerged as an important sequence for the analysis of cerebrovascular (CVD) and Alzheimer's disease (AD). Large-scale, automated cross-sectional and longitudinal cerebral biomarker extraction from FLAIR datasets could progress disease characterization, improve disease monitoring, and help to determine optimal intervention times. Despite this, most automated biomarker extraction algorithms are designed for T1-weighted or multi-modal inputs. In this work, automated tools were used to extract biomarkers from large, FLAIR-only datasets to evaluate the feasibility of this sequence to characterize healthy, AD, and CVD subjects in a similar manner to traditional approaches. Total brain volume (TBV), cerebrospinal fluid (CSF) volume, and white matter lesion (WML) volume was measured for the cross-sectional biomarkers and the corresponding annual rates of change over multiple scans represented the longitudinal biomarkers. Biomarkers were extracted from two dementia cohorts (4356 vol, 162 233 images) and one vascular disease cohort (869 vol, 42 850 images) using deep learning-based segmentation algorithms designed specifically for FLAIR. Biomarkers from all cohorts were summarized using descriptive statistics, correlation analysis, and ANCOVA to assess differences in diagnostic labels while accounting for demographic and acquisition factors. Biomarkers from FLAIR MRI had similar trends with those extracted from traditional modalities in the literature for characterizing healthy, AD, and CVD subjects. This demonstrates that FLAIR MRI can be used for end-to-end analysis of large AD and CVD datasets, which can lower acquisition costs, simplify clinical translation, and reduce measurement error associated with multi-modal approaches.

Semantic segmentation of multispectral photoacoustic images using deep learning

Photoacoustic (PA) imaging has the potential to revolutionize functional medical imaging in healthcare due to the valuable information on tissue physiology contained in multispectral photoacoustic measurements. Clinical translation of the technology requires conversion of the high-dimensional acquired data into clinically relevant and interpretable information. In this work, we present a deep learning-based approach to semantic segmentation of multispectral photoacoustic images to facilitate image interpretability. Manually annotated photoacoustic and ultrasound imaging data are used as reference and enable the training of a deep learning-based segmentation algorithm in a supervised manner. Based on a validation study with experimentally acquired data from 16 healthy human volunteers, we show that automatic tissue segmentation can be used to create powerful analyses and visualizations of multispectral photoacoustic images. Due to the intuitive representation of high-dimensional information, such a preprocessing algorithm could be a valuable means to facilitate the clinical translation of photoacoustic imaging.

Open-Source Automatic Biomarker Measurement on Slit-Lamp Photography to Estimate Visual Acuity in Microbial Keratitis.

PurposeTo assess clinical applicability of automatic image analysis in microbial keratitis (MK) by evaluating the relationship between biomarker measurements on slit-lamp photography (SLP) and best-corrected visual acuity (BCVA).MethodsSeventy-six patients with MK with SLP images and same-day logarithm of the minimum angle of resolution (logMAR) BCVA were evaluated. MK biomarkers (stromal infiltrate, white blood cell infiltration, corneal edema, hypopyon, epithelial defect) were segmented manually by ophthalmologists and automatically by a novel, open-source, deep learning–based segmentation algorithm. Five measurements (presence, maximum width, total area, proportion of the corneal limbus area affected, centrality) were calculated. Correlations between the measurements and BCVA were calculated. An automatic regression model estimated BCVA from the measurements. Differences in performance between using manual and automatic measurements were evaluated using William's test (for correlation) and the paired-sample t-test (for absolute error).ResultsMeasurements had high correlations of 0.86 (manual) and 0.84 (automatic) with true BCVA. Estimated BCVA had average (mean ± SD) absolute errors of 0.39 ± 0.27 logMAR (manual, median: 0.30) and 0.35 ± 0.28 logMAR (automatic, median: 0.30) and high correlations of 0.76 (manual) and 0.80 (automatic) with true BCVA. Differences between using manual and automatic measurements were not statistically significant for correlations of measurements with true BCVA (P = .66), absolute errors of estimated BCVA (P = .15), or correlations of estimated BCVA with true BCVA (P = .60).ConclusionsThe proposed algorithm measured MK biomarkers as accurately as ophthalmologists. Measurements were highly correlated with and estimative of visual acuity.Translational RelevanceThis study demonstrates the potential of developing fully automatic objective and standardized strategies to aid ophthalmologists in the clinical assessment of MK.

Abstract P343: Relationship of Infarct Volume to Inflammation and Prognostic Significance in Patients With Acute Ischemic Stroke

Background and Purpose: The diffusion weighted imaging (DWI) lesion volumes in acute ischemic stroke (AIS) can be automatically measured using deep learning-based segmentation algorithms. We aim to explore the prognostic significance of artificial intelligence-predicted infarct volume, and the association of markers of acute inflammation with the infarct volume. Methods: 12,598 AIS/TIA patients were included in this analysis. Intarct volume was automatically measured using a U-Net model for acute ischemic stroke lesion segmentation on DWI. Participants were divided into 5 subgroups according to infarct volume. Spearman’s correlations were employed to study the association between infarct volume and markers of acute inflammation. Multivariable logistic regression and Cox proportional hazards model were performed to explore the relationship between infarct volume and the incidence of poor functional outcome (modified Rankin scale score 3-6), stroke recurrence or combined vascular events at 3 months. Results: The U-Net model prediction correlated and agreed well with manual annotation ground truth for infarct volume (r=0.96; P&lt;0.001). There were positive correlations between the infarct volume and markers of acute inflammation (neutrophil [r=0.175; P&lt;0.001], hs-CRP [r=0.180; P&lt;0.001], and IL-6 [r=0.225; P&lt;0.001]). Compared with those without DWI lesions, patients with the largest infarct volume (4th Quartile) were nearly five times more likely to have poor functional outcome (mRS 3-6) (adjusted odds ratio, 4.70; 95% confidence intervals [CI], 3.29-6.72; P for trend&lt;0.001) after adjustment for confounding factors and markers of acute inflammation. The infarct volume category was significantly associated with stroke recurrence (adjusted hazard ratios [HRs], 1.0, 1.43[0.95,2.17], 2.22[1.49,3.29], 2.06[1.40,3.05], 2.26[1.52,3.36]; P for trend&lt;0.001) and combined vascular events(adjusted HRs, 1.0, 1.38[0.92,2.09], 2.25[1.53,3.32], 2.03[1.38,2.98], 2.28[1.54,3.36]; P for trend&lt;0.001). Conclusions: Infarct volume measured automatically by deep learning-based tool was a strong predictor of poor functional outcome as well as stroke recurrence, with the potential for widespread adoption in both research and clinical settings.

Automatic segmentation of pulmonary lobes on low-dose computed tomography using deep learning.

BackgroundTo develop and validate a fully automated deep learning-based segmentation algorithm to segment pulmonary lobe on low-dose computed tomography (LDCT) images.MethodsThis study presents an automatic segmentation of pulmonary lobes using a fully convolutional neural network named dense V-network (DenseVNet) on lung cancer screening LDCT images. A total of 160 LDCT cases for lung cancer screening (100 in the training set, 10 in the validation set, and 50 in the test set) was included in this study. Specifically, the template of pulmonary lobes (the right lung consists of three lobes, and the left lung consists of two lobes) were obtained from pixel-level annotations by semiautomatic segmentation platform. Then, the model was trained under the supervision of the LDCT training set. Finally, the trained model was used to segment the LDCT in the test set. Dice coefficient, Jaccard coefficient, and Hausdorff distance were adopted as evaluation metrics to verify the performance of our segmentation model.ResultsIn this study, the model achieved the accurate segmentation of each pulmonary lobe in seconds without the intervention of researchers. The testing set consisted 50 LDCT cases were used to evaluate the performance of the segmentation model. The all-lobes Dice coefficient of the test set was 0.944, the Jaccard coefficient was 0.896, and the Hausdorff distance was 92.908 mm.ConclusionsThe segmentation model based on LDCT can automatically and robustly and efficiently segment pulmonary lobes. It will provide effective location information and contour constraints for pulmonary nodule detection on LDCT images for lung cancer screening, which may have potential clinical application.

Accurate and Automatic 3D Segmentation of Femur and Pelvis from CT Images of the Hip based on Deep Learning

Background: Prosthesis design of hip joint and computer-assisted surgical planning can benefit from segmentation-based computed tomography (CT). Purpose: To automatically segment the three-dimensional (3D) hip joint images, the authors developed a deep learning-based segmentation algorithm and verified its feasibility with CT image of hip joint. Methods: Conventional image augmentation strategies and specific image augmentation strategies, which were designed to mimic the deformed shape and blurring boundaries of the diseased hips, were applied to obtain a large number of training samples of diseased hips to avoid overfitting. A 3D segmentation algorithm named light 3D U-net, which segmented images from coarse to fine, was developed to improve the segmentation accuracy and reduce the computation time in 3D hip joint image with multiple targets. Results and Discussion: The test results showed that the proposed method would exceed 0.9 in the aspects of Dice score, Specificity, and Sensitivity. Comparing with traditional method, the proposed method is more efficient, accurate, and robust. The proposed method has shown great potential to be applied in prosthesis design and computer-assisted surgical planning.

Evaluation of image segmentation techniques for image-based rock property estimation

Accurate characterization of rock and fluid properties in porous media using X-ray imaging techniques depends on reliable identification and segmentation of the involved phases. Segmentation is critical for the estimation of porosity, fluid saturations, fluid and rock topology, and pore connectivity, among other pore-scale properties. Therefore, the purpose of this study is to compare the effectiveness of different image segmentation techniques when applied to image data analysis in porous media. Two machine learning based segmentation techniques – a supervised ML technique called Fast Random Forest, and an unsupervised method combining k-means and fuzzy c-means clustering algorithms – were compared using an experimental data set. Comparisons are also presented against traditional thresholding segmentation. In addition, we discuss the potential and limitations of applying deep learning based segmentation algorithms. The performance of the segmentation techniques were compared on estimates of porosity, saturation, and surface area, as well as pore-scale estimates such as fluid-fluid interfacial areas, and Euler characteristic. X-ray micro-computed tomography images for a sintered glass frit, saturated with two-phases (air and brine), were acquired at two different voxel resolutions. The high-resolution images (6 μm) were used as the benchmark case, while the low-resolution images (18 μm) were segmented by three segmentation techniques: Fast Random Forest, clustering, and thresholding. The results for porosity and phase saturation from thresholding and from the supervised ML method (i.e. Fast Random Forest) were found to be close to the benchmark case. Segmentation results from the unsupervised ML method (i.e. clustering) were largely unsatisfactory, except for total surface area measurements. The supervised ML segmentation results provided better measurements for air-brine interfacial areas by capturing three-phase interfacial regions. Also, all segmentation techniques resulted in similar measurements for air-phase Euler characteristic confirming poor connectivity of the trapped air phase, although the closest results were obtained by the supervised ML method. Finally, despite the supervised ML segmentation technique being more computationally intensive, it was found to require less user intervention and its implementation was more straightforward. In summary, this work provides insights into different segmentation techniques, their implementation, as well as advantages and limitations with regards to quantitative analysis of pore-scale properties in saturated porous media.

Benchmarking algorithms for food localization and semantic segmentation

The problem of food segmentation is quite challenging since food is characterized by intrinsic high intra-class variability. Also, segmentation of food images taken in-the-wild may be characterized by acquisition artifacts, and that could be problematic for the segmentation algorithms. A proper evaluating of segmentation algorithms is of paramount importance for the design and improvement of food analysis systems that can work in less-than-ideal real scenarios. In this paper, we evaluate the performance of different deep learning-based segmentation algorithms in the context of food. Due to the lack of large-scale food segmentation datasets, we initially create a new dataset composed of 5000 images of 50 diverse food categories. The images are accurately annotated with pixel-wise annotations. In order to test the algorithms under different conditions, the dataset is augmented with the same images but rendered under different acquisition distortions that comprise illuminant change, JPEG compression, Gaussian noise, and Gaussian blur. The final dataset is composed of 120,000 images. Using standard benchmark measures, we conducted extensive experiments to evaluate ten state-of-the-art segmentation algorithms on two tasks: food localization and semantic food segmentation.

A web-based brain metastases segmentation and labeling platform for stereotactic radiosurgery.

Stereotactic radiosurgery (SRS) has becomea standard of care for patients' withbrain metastases (BMs). However, the manualmultipleBMs delineation can be time-consuming and could create an efficiencybottleneck in SRS workflow. There is a clinical needfor automatic delineation and quantitative evaluationtools. In this study, building on our previous developed deep learning-based segmentation algorithms, we developed a web-based automated BMs segmentation and labeling platform to assist the SRS clinical workflow. This platformwas developed based on the Django framework, including a web client and a back-end server. The web client enables interactions as database access, data import, andimage viewing. The server performs the segmentation and labeling tasks including: skull stripping; deep learning-based BMs segmentation; and affine registration-based BMs labeling. Additionally, the client can display BMs contours with corresponding atlas labels, and allows further postprocessing tasks including: (a) adjusting window levels; (b) displaying/hiding specific contours; (c) removing false-positive contours; (d) exporting contours as DICOM RTStruct files; etc. RESULTS: We evaluated this platform on 10 clinical cases with BMs number varied from 12-81per case. The overall operation took about 4-5min per patient. The segmentation accuracywas evaluated between the manual contour and automatic segmentation withseveral metrics.The averaged center of mass shift was 1.55±0.36mm, theHausdorff distance was 2.98±0.63mm, the mean of surface-to-surface distance (SSD) was 1.06±0.31mm, andthe standard deviation of SSD was 0.80±0.16mm. In addition, the initial averaged false-positive over union (FPoU) and false-negative rate (FNR) were 0.43±0.19 and 0.15±0.10 respectively. After case-specific postprocessing, the averaged FPoU and FNR were 0.19±0.10 and 0.15±0.10 respectively. The evaluated web-based BMs segmentation and labeling platform can substantially improve the clinical efficiencycompared to manual contouring. This platform can be a useful tool for assisting SRS treatment planning and treatment follow-up.

Beyond Performance Metrics: Automatic Deep Learning Retinal OCT Analysis Reproduces Clinical Trial Outcome

To validate the efficacy of a fully automatic, deep learning-based segmentation algorithm beyond conventional performance metrics by measuring the primary outcome of a clinical trial for macular telangiectasia type 2 (MacTel2). Evaluation of diagnostic test or technology. A total of 92 eyes from 62 participants with MacTel2 from a phase 2 clinical trial (NCT01949324) randomized to 1 of 2 treatment groups METHODS: The ellipsoid zone (EZ) defect areas were measured on spectral domain OCT images of each eye at 2 time points (baseline and month 24) by a fully automatic, deep learning-based segmentation algorithm. The change in EZ defect area from baseline to month 24 was calculated and analyzed according to the clinical trial protocol. Difference in the change in EZ defect area from baseline to month 24 between the 2 treatment groups. The difference in the change in EZ defect area from baseline to month 24 between the 2 treatment groups measured by the fully automatic segmentation algorithm was 0.072±0.035 mm2 (P= 0.021). This was comparable to the outcome of the clinical trial using semiautomatic measurements by expert readers, 0.065±0.033 mm2 (P= 0.025). The fully automatic segmentation algorithm was as accurate as semiautomatic expert segmentation to assess EZ defect areas and was able to reliably reproduce the statistically significant primary outcome measure of the clinical trial. This approach, to validate the performance of an automatic segmentation algorithm on the primary clinical trial end point, provides a robust gauge of its clinical applicability.

Pupil diameter in migraine and tension headache.

Pupil diameter was measured in darkness and in dull and bright illumination in 39 migrainous patients and in 15 tension headache sufferers during headache. In 21 migrainous patients, measurements were repeated during the headache-free interval. Mean pupil diameter was smaller in patients with common migraine at the time of examination than in 20 nonheadache control subjects, and smaller on the symptomatic side in migrainous patients with unilateral headache. During the headache-free interval mean pupil diameter did not differ from values in non-headache controls. These findings suggest that migraine is associated with a sympathetic pupillary deficit which is greater on the habitually-affected side.