Automatic Segmentation Research Articles

Segmentation and characterization of macerated fibers and vessels using deep learning

PurposeWood comprises different cell types, such as fibers, tracheids and vessels, defining its properties. Studying cells’ shape, size, and arrangement in microscopy images is crucial for understanding wood characteristics. Typically, this involves macerating (soaking) samples in a solution to separate cells, then spreading them on slides for imaging with a microscope that covers a wide area, capturing thousands of cells. However, these cells often cluster and overlap in images, making the segmentation difficult and time-consuming using standard image-processing methods.ResultsIn this work, we developed an automatic deep learning segmentation approach that utilizes the one-stage YOLOv8 model for fast and accurate segmentation and characterization of macerated fiber and vessel form aspen trees in microscopy images. The model can analyze 32,640 x 25,920 pixels images and demonstrate effective cell detection and segmentation, achieving a mAP0.5-0.95\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\hbox {mAP}_{0.5-0.95}$$\\end{document} of 78 %. To assess the model’s robustness, we examined fibers from a genetically modified tree line known for longer fibers. The outcomes were comparable to previous manual measurements. Additionally, we created a user-friendly web application for image analysis and provided the code for use on Google Colab.ConclusionBy leveraging YOLOv8’s advances, this work provides a deep learning solution to enable efficient quantification and analysis of wood cells suitable for practical applications.

Automatic deep learning segmentation of the hippocampus on high-resolution diffusion magnetic resonance imaging and its application to the healthy lifespan.

Diffusion tensor imaging (DTI) can provide unique contrast and insight into microstructural changes with age or disease of the hippocampus, although it is difficult to measure the hippocampus because of its comparatively small size, location, and shape. This has been markedly improved by the advent of a clinically feasible 1-mm isotropic resolution 6-min DTI protocol at 3 T of the hippocampus with limited brain coverage of 20 axial-oblique slices aligned along its long axis. However, manual segmentation is too laborious for large population studies, and it cannot be automatically segmented directly on the diffusion images using traditional T1 or T2 image-based methods because of the limited brain coverage and different contrast. An automatic method is proposed here that segments the hippocampus directly on high-resolution diffusion images based on an extension of well-known deep learning architectures like UNet and UNet++ by including additional dense residual connections. The method was trained on 100 healthy participants with previously performed manual segmentation on the 1-mm DTI, then evaluated on typical healthy participants (n = 53), yielding an excellent voxel overlap with a Dice score of ~ 0.90 with manual segmentation; notably, this was comparable with the inter-rater reliability of manually delineating the hippocampus on diffusion magnetic resonance imaging (MRI) (Dice score of 0.86). This method also generalized to a different DTI protocol with 36% fewer acquisitions. It was further validated by showing similar age trajectories of volumes, fractional anisotropy, and mean diffusivity from manual segmentations in one cohort (n = 153, age 5-74 years) with automatic segmentations from a second cohort without manual segmentations (n = 354, age 5-90 years). Automated high-resolution diffusion MRI segmentation of the hippocampus will facilitate large cohort analyses and, in future research, needs to be evaluated on patient groups.

Research Note: A deep learning method segments chicken keel bones from whole-body X-ray images

Most commercial laying hens suffer from sternum (keel) bone damage including deviations and fractures. X-raying hens, followed by segmenting and assessing the keel bone, is a key to automating the monitoring of keel bone condition. The aim of the current work is to train a deep learning model to segment the keel bone out of whole-body x-ray images. We obtained full-body x-ray images of laying hens (n = 1,051) and manually drew the outline of the keel bone on each image. Using the annotated images, a U-net model was then trained to segment the keel bone. The proposed model was evaluated using 5-fold cross validation. We obtained high segmentation accuracy (Dice coefficients of 0.88-0.90) repeatably over several validation folds. In conclusion, automatic segmentation of the keel bone from full-body x-ray images is possible with good accuracy. Segmentation is a requirement for automated measurements of keel geometry and density, which can subsequently be connected to susceptibility to keel deviations and fractures.

Comparative investigation on the performance of spot positioning sensors in point-autofocus microscopy

As products evolve towards greater precision and intelligence, their components also develop towards miniaturisation and increasing complexity. In the era of intelligence, the demand for complex small components continues to grow. The point-autofocus microscopy (PAM), with its attributes such as small measurement spots and high vertical resolution, holds a comprehensive advantage in instruments for measuring the geometry of complex small components. The performance of the spot positioning sensor directly determines the capability of the PAM to measure complex small components. This study introduces the basic measurement principles of the PAM and commonly used spot positioning sensors. The analysis explored the feasibility of applying different sensors to the PAM. An automatic segmentation and weighted localisation algorithm of the spot based on the centroid method was proposed. Using this algorithm, the spot positioning error of the complementary metal–oxide–semiconductor (CMOS) sensor can be controlled within 0.48 μm, with a measurement uncertainty of no more than ±0.19 μm. This study focuses on comparing the performance of CMOS and position-sensitive detector sensors with the performance of seven indicators. In addition, a measurement system of the PAM was constructed, and measurements of the test object were conducted using both types of sensors. The experimental results show that when the measurement system employs the CMOS sensor, it achieves higher measurement accuracy. The maximum measurement error was approximately 0.22 μm, with a measurement uncertainty of no more than ±0.10 μm.

Retraction Note: Automatic segmentation and classification of liver tumor from CT image using feature difference and SVM based classifier-soft computing technique

Retraction Note: Automatic segmentation and classification of liver tumor from CT image using feature difference and SVM based classifier-soft computing technique

Automated Cerebrovascular Segmentation and Visualization of Intracranial Time-of-Flight Magnetic Resonance Angiography Based on Deep Learning.

Time-of-flight magnetic resonance angiography (TOF-MRA) is a non-contrast technique used to visualize neurovascular. However, manual reconstruction of the volume render (VR) by radiologists is time-consuming and labor-intensive. Deep learning-based (DL-based) vessel segmentation technology may provide intelligent automation workflow. To evaluate the image quality of DL vessel segmentation for automatically acquiring intracranial arteries in TOF-MRA. A total of 394 TOF-MRA scans were selected, which included cerebral vascular health, aneurysms, or stenoses. Both our proposed method and two state-of-the-art DL methods are evaluated on external datasets for generalization ability. For qualitative assessment, two experienced clinical radiologists evaluated the image quality of cerebrovascular diagnostic and visualization (scoring 0-5 as unacceptable to excellent) obtained by manual VR reconstruction or automatic convolutional neural network (CNN) segmentation. The proposed CNN outperforms the other two DL-based methods in clinical scoring on external datasets, and its visualization was evaluated by readers as having the appearance of the radiologists' manual reconstructions. Scoring of proposed CNN and VR of intracranial arteries demonstrated good to excellent agreement with no significant differences (median, 5.0 and 5.0, P ≥ 12) at healthy-type scans. All proposed CNN image quality were considered to have adequate diagnostic quality (median scores > 2). Quantitative analysis demonstrated a superior dice similarity coefficient of cerebrovascular overlap (training sets and validation sets; 0.947 and 0.927). Automatic cerebrovascular segmentation using DL is feasible and the image quality in terms of vessel integrity, collateral circulation and lesion morphology is comparable to expert manual VR without significant differences.

Automatic segmentation of dental cone-beam computed tomography scans using a deep learning framework

Bevezetés: A ’cone-beam’ (kúpsugaras) számítógépes tomográfiás (CBCT) felvételek szegmentációja során a síkbeli képekből álló adatokat három dimenzióban (3D) rekonstruáljuk. A szájsebészetben és a parodontológiában a digitális adatfeldolgozás lehetővé teszi a műtéti beavatkozások 3D tervezését. A leggyakrabban alkalmazott határérték-alapú szegmentáció gyors, de pontatlan, míg a félautomatikus módszerek megfelelő pontosságúak, de rendkívül időigényesek. Az utóbbi években a mesterséges intelligencián alapuló technológiák elterjedésével azonban mostanra lehetőség van a CBCT-felvételek automatikus szegmentációjára. Célkitűzés: A klinikai gyakorlatból vett CBCT-felvételeken betanított mélytanulási szegmentációs modell bemutatása és hatékonyságának vizsgálata. Módszer: A vizsgálat három fő fázisa volt: a tanuló adatbázis felállítása, a mélytanulási modell betanítása és ezen architektúra pontosságának tesztelése. A tanuló adatbázis felállításához 70, részlegesen fogatlan páciens CBCT-felvételeit alkalmaztuk. A SegResNet hálózati architektúrára épülő szegmentációs modellt a MONAI rendszer segítségével fejlesztettük ki. A mélytanulási modell pontosságának ellenőrzéséhez 15 CBCT-felvételt használtunk. Ezeket a felvételeket a mélytanulási modell segítségével, valamint félautomatikus szegmentációval is feldolgoztuk, és összehasonlítottuk a két szegmentáció eredményét. Eredmények: A mélytanulásos szegmentáció és a félautomatikus szegmentáció közötti hasonlóság a Jaccard-index szerint átlagosan 0,91 ± 0,02, a Dice hasonlósági együttható átlagos értéke 0,95 ± 0,01, míg a két modell közötti átlagos Hausdorff- (95%) távolság 0,67 mm ± 0,22 mm volt. A mélytanulásos architektúra által szegmentált és a félautomatikus szegmentációval létrehozott 3D modellek térfogata nem mutatott statisztikailag szignifikáns különbséget (p = 0,31). Megbeszélés: A vizsgálatunkban használt mélytanulási modell az irodalomban található mesterségesintelligencia-rendszerekhez hasonló pontossággal végezte el a CBCT-felvételek szegmentációját, és mivel a CBCT-felvételek a rutin klinikai gyakorlatból származtak, a mélytanulási modell relatíve nagy megbízhatósággal szegmentálta a parodontalis csonttopográfiát és az alveolaris gerincdefektusokat. Következtetés: A mélytanulási modell nagy pontossággal szegmentálta az alsó állcsontot dentális CBCT-felvételeken. Ezek alapján megállapítható, hogy a mélytanulásos szegmentációval előállított 3D modell alkalmas lehet rekonstruktív szájsebészeti és parodontalis sebészeti beavatkozások digitális tervezésére. Orv Hetil. 2024; 165(32): 1242–1251.

Radiomics signature for automatic hydronephrosis detection in unenhanced Low-Dose CT

PurposeTo investigate the diagnostic performance of an automatic pipeline for detection of hydronephrosis on kidney's parenchyma on unenhanced low-dose CT of the abdomen. MethodsThis retrospective study included 95 patients with confirmed unilateral hydronephrosis in an unenhanced low-dose CT of the abdomen. Data were split into training (n = 67) and test (n = 28) cohorts. Both kidneys for each case were included in further analyses, whereas the kidney without hydronephrosis was used as control. Using the training cohort, we developed a pipeline consisting of a deep-learning model for automatic segmentation (a Convolutional Neural Network based on nnU-Net architecture) of the kidney's parenchyma and a radiomics classifier to detect hydronephrosis. The models were assessed using standard classification metrics, such as area under the ROC curve (AUC), sensitivity and specificity, as well as semantic segmentation metrics, including Dice coefficient and Jaccard index. ResultsUsing manual segmentation of the kidney’s parenchyma, hydronephrosis can be detected with an AUC of 0.84, a sensitivity of 75% and a specificity of 82%, a PPV of 81% and a NPV of 77%. Automatic kidney segmentation achieved a mean Dice score of 0.87 and 0.91 for the right and left kidney, respectively. Additionally, automatic segmentation achieved an AUC of 0.83, a sensitivity of 86%, specificity of 64%, PPV of 71%, and NPV of 82%. ConclusionOur proposed radiomics signature using automatic kidney's parenchyma segmentation allows for accurate hydronephrosis detection on unenhanced low-dose CT scans of the abdomen independently of widened renal pelvis. This method could be used in clinical routine to highlight hydronephrosis to radiologists as well as clinicians, especially in patients with concurrent parapelvic cysts and might reduce time and costs associated with diagnosing hydronephrosis.

Accurate neuron segmentation method for one-photon calcium imaging videos combining convolutional neural networks and clustering

One-photon fluorescent calcium imaging helps understand brain functions by recording large-scale neural activities in freely moving animals. Automatic, fast, and accurate active neuron segmentation algorithms are essential to extract and interpret information from these videos. One-photon imaging videos’ low resolution, high noise, and high background fluctuation pose significant challenges. Here, we develop a software pipeline to address the challenges of processing one-photon calcium imaging videos. We extend our previous two-photon active neuron segmentation algorithm, Shallow U-Net Neuron Segmentation (SUNS), to better suppress background fluctuations in one-photon videos. We also develop additional neuron extraction (ANE) to locate small or dim neurons missed by SUNS. To train our segmentation method, we create ground truth neurons by developing a manual labeling pipeline assisted with semi-automatic refinement. Our method is more accurate and faster than state-of-the-art techniques when processing simulated videos and multiple experimental datasets acquired over various brain regions with different imaging conditions.

Identifying sources of bias when testing three available algorithms for quantifying white matter lesions: BIANCA, LPA and LGA.

Brain magnetic resonance imaging frequently reveals white matter lesions (WMLs) in older adults. They are often associated with cognitive impairment and risk of dementia. Given the continuous search for the optimal segmentation algorithm, we broke down this question by exploring whether the output of algorithms frequently used might be biased by the presence of different influencing factors. We studied the impact of age, sex, blood glucose levels, diabetes, systolic blood pressure and hypertension on automatic WML segmentation algorithms. We evaluated three widely used algorithms (BIANCA, LPA and LGA) using the population-based 1000BRAINS cohort (N = 1166, aged 18-87, 523 females, 643 males). We analysed two main aspects. Firstly, we examined whether training data (TD) characteristics influenced WML estimations, assessing the impact of relevant factors in the TD. Secondly, algorithm's output and performance within selected subgroups defined by these factors were assessed. Results revealed that BIANCA's WML estimations are influenced by the characteristics present in the TD. LPA and LGA consistently provided lower WML estimations compared to BIANCA's output when tested on participants under 67 years of age without risk cardiovascular factors. Notably, LPA and LGA showed reduced accuracy for these participants. However, LPA and LGA showed better performance for older participants presenting cardiovascular risk factors. Results suggest that incorporating comprehensive cohort factors like diverse age, sex and participants with and without hypertension in the TD could enhance WML-based analyses and mitigate potential sources of bias. LPA and LGA are a fast and valid option for older participants with cardiovascular risk factors.

VMseg: Using spatial variance to automatically segment retinal non-perfusion on OCT-angiography.

To develop and test VMseg, a new image processing algorithm performing automatic segmentation of retinal non-perfusion in widefield OCT-Angiography images, in order to estimate the non-perfusion index in diabetic patients. We included diabetic patients with severe non-proliferative or proliferative diabetic retinopathy. We acquired images using the PlexElite 9000 OCT-A device with a photomontage of 5 images of size 12 x 12 mm. We then developed VMseg, a Python algorithm for non-perfusion detection, which binarizes a variance map calculated through convolution and morphological operations. We used 70% of our data set (development set) to fine-tune the algorithm parameters (convolution and morphological parameters, binarization thresholds) and evaluated the algorithm performance on the remaining 30% (test set). The obtained automatic segmentations were compared to a ground truth corresponding to manual segmentation from a retina expert and the inference processing time was estimated. We included 51 eyes of 30 patients (27 severe non-proliferative, 24 proliferative diabetic retinopathy). Using the optimal parameters found on the development set to tune the algorithm, the mean dice for the test set was 0.683 (sd = 0.175). We found a higher dice coefficient for images with a higher area of retinal non-perfusion (rs = 0.722, p < 10-4). There was a strong correlation (rs = 0.877, p < 10-4) between VMseg estimated non-perfusion indexes and indexes estimated using the ground truth segmentation. The Bland-Altman plot revealed that 3 eyes (5.9%) were significantly under-segmented by VMseg. We developed VMseg, an automatic algorithm for retinal non-perfusion segmentation on 12 x 12 mm OCT-A widefield photomontages. This simple algorithm was fast at inference time, segmented images in full-resolution and for the OCT-A format, was accurate enough for automatic estimation of retinal non-perfusion index in diabetic patients with diabetic retinopathy.

Artificial Intelligence Measurement of Preoperative Radiographs in Adolescent Idiopathic Scoliosis Based on Multiple-View Semantic Segmentation.

Cross-sectional study. Imaging classification of adolescent idiopathic scoliosis (AIS) is directly related to the surgical strategy, but the artificial classification is complex and depends on doctors' experience. This study investigated deep learning-based automated classification methods (DL group) for AIS and validated the consistency of machine classification and manual classification (M group). A total of 506 cases (81 males and 425 females) and 1812 AIS full spine images in the anteroposterior (AP), lateral (LAT), left bending (LB) and right bending (RB) positions were retrospectively used for training. The mean age was 13.6 ± 1.8. The mean maximum Cobb angle was 46.8 ± 12.0. U-Net semantic segmentation neural network technology and deep learning methods were used to automatically segment and establish the alignment relationship between multiple views of the spine, and to extract spinal features such as the Cobb angle. The type of each test case was automatically calculated according to Lenke's rule. An additional 107 cases of adolescent idiopathic scoliosis imaging were prospectively used for testing. The consistency of the DL group and M group was compared. Automatic vertebral body segmentation and recognition, multi-view alignment of the spine and automatic Cobb angle measurement were implemented. Compare to the M group, the consistency of the DL group was significantly higher in 3 aspects: type of lateral convexity (0.989 vs 0.566), lumbar curvature modifier (0.932 vs 0.738), and sagittal plane modifier (0.987 vs 0.522). Deep learning enables automated Cobb angle measurement and automated Lenke classification of idiopathic scoliosis whole spine radiographs with higher consistency than manual measurement classification.

Improved building MEP systems semantic segmentation in point clouds using a novel multi-class dataset and local–global vector transformer network

Point cloud semantic segmentation for mechanical, electrical, and plumbing (MEP) systems is crucial for establishing MEP systems digital twins. Deep learning has shown promise in automatically deriving point-wise semantic labels from point clouds. However, the utilization of 3D deep learning methods for directly segmenting complex, occluded, and densely distributed multi-class MEP systems in common building point clouds remains unexplored due to the lack of specialized public datasets. To address this, our study introduces a public, richly annotated, and multi-class point cloud dataset for MEP systems in common buildings, consisting of 92.10 million points across four buildings, 56 areas, and nine component categories. Subsequently, we propose Trans2Net, a novel and customized 3D deep learning method for the automatic and precise segmentation of MEP components, featuring local vector transformer module, global-aid vector transformer module, and learnable feature fusion operator that enhance the model’s sensitivity to local features, capture global context with sparse points, and dynamically fuse weighted local and global features. Experimental results demonstrate that Trans2Net significantly outperforms seven other state-of-the-art methods on the MEP dataset, achieving an overall accuracy of 97.39%, a mean accuracy of 95.45%, and a mean intersection over union (mIoU) of 90.76%. Notably, it surpasses the second-ranked method, PointMeta, by a margin of 9.91% in mIoU. This study provides a novel paradigm for automatically understanding MEP systems in point clouds and offers robust support for digital twin management of MEP systems, contributing to the prolonged lifespans of MEP systems.

Deep Machine Learning for Automatic Segmentation of the Pancreatic Parenchyma and its hypo- and hypervascular lesions on CT Images

Objective of the study. To develop and evaluate the effectiveness of a technology for segmenting the pancreatic parenchyma and its hyper- and hypovascular lesions on abdominal computed tomography (CT) scans using deep machine learning.Materials and methods. CT scans from the database of the A.V. Vishnevsky National Medical Research Center of Surgery were used for training and testing the algorithms – a total number of approximately 150 studies (arterial and venous phases). A test dataset of 46 anonymized CT scans (arterial and venous phases) was prepared for validation of the obtained algorithms, independently assessed by expert physicians. The primary segmentation neural network used is nn-UNet (M. Antonelli et al., 2022).Results. The average accuracy of the test dataset for the model determining segmentation masks of the pancreas on CT images had an AUC of 0.8 for the venous phase and 0.85 for the arterial phase. The segmentation masks of pancreatic formations had an AUC of 0.6.Conclusion. Automated segmentation of the pancreatic parenchyma structure using deep machine learning technologies demonstrated high accuracy. However, the segmentation of hypo- and hypervascular pancreatic lesions requires improvement. The overlap of the masks showed a rather low result, but in all cases, the location of the pathological formation was correctly identified by the algorithm. Enhancing the training dataset and the algorithm used could increase the accuracy of the algorithm.No false negative results were obtained when detecting pancreatic formations; in all cases, the INS detected “suspicious” areas of the pancreatic parenchyma. This can help reduce the omission of pancreatic pathologies in CT scans, and their further assessment can be carried out by the radiologist himself.

AI-based automated segmentation for ovarian/adnexal masses and their internal components on ultrasound imaging.

Segmentation of ovarian/adnexal masses from surrounding tissue on ultrasound images is a challenging task. The separation of masses into different components may also be important for radiomic feature extraction. Our study aimed to develop an artificial intelligence-based automatic segmentation method for transvaginal ultrasound images that (1)outlines the exterior boundary of adnexal masses and (2)separates internal components. A retrospective ultrasound imaging database of adnexal masses was reviewed for exclusion criteria at the patient, mass, and image levels, with one image per mass. The resulting 54 adnexal masses (36 benign/18 malignant) from 53 patients were separated by patient into training (26 benign/12 malignant) and independent test (10 benign/6 malignant) sets. U-net segmentation performance on test images compared to expert detailed outlines was measured using the Dice similarity coefficient (DSC) and the ratio of the Hausdorff distance to the effective diameter of the outline ( ) for each mass. Subsequently, in discovery mode, a two-level fuzzy c-means (FCM) unsupervised clustering approach was used to separate the pixels within masses belonging to hypoechoic or hyperechoic components. The DSC (median [95% confidence interval]) was 0.91 [0.78, 0.96], and was 0.04 [0.01, 0.12], indicating strong agreement with expert outlines. Clinical review of the internal separation of masses into echogenic components demonstrated a strong association with mass characteristics. A combined U-net and FCM algorithm for automatic segmentation of adnexal masses and their internal components achieved excellent results compared with expert outlines and review, supporting future radiomic feature-based classification of the masses by components.

Detection of Brain Tumor Employing Residual Network-based Optimized Deep Learning

Diagnosis and treatment planning play a very vital role in improving the survival of oncological patients. However, there is high variability in the shape, size, and structure of the tumor, making automatic segmentation difficult. The automatic and accurate detection and segmentation methods for Brain tumors are proposed in this paper. A modified ResNet50 model was used for tumor detection, and a ResUNetmodel-based convolutional neural network for segmentation is proposed in this paper. The detection and segmentation were performed on the same dataset consisting of pre-contrast, FLAIR, and postcontrast MRI images of 110 patients collected from the Cancer Imaging Archive. Due to the use of Residual Networks, the authors observed improvement in evaluation parameters, such as accuracy for tumor detection and dice similarity coefficient for tumor segmentation. The accuracy of tumor detection and Dice Similarity Coefficient achieved by the segmentation model were 96.77% and 0.893, respectively, for the TCIA dataset. The results were compared based on manual segmentation and existing segmentation techniques. The tumor mask was also individually compared to the ground truth using the SSIM value. The proposed detection and segmentation models were validated on BraTS2015 and BraTS2017 datasets, and the results were consensus. The use of residual networks in both the detection and the segmentation model resulted in improved accuracy and DSC score. DSC score was increased by 5.9% compared to the UNet model, and the accuracy of the model was increased from 92% to 96.77% for the test set.

Interactive Cascaded Network for Prostate Cancer Segmentation from Multimodality MRI with Automated Quality Assessment.

The accurate segmentation of prostate cancer (PCa) from multiparametric MRI is crucial in clinical practice for guiding biopsy and treatment planning. Existing automated methods often lack the necessary accuracy and robustness in localizing PCa, whereas interactive segmentation methods, although more accurate, require user intervention on each input image, thereby limiting the cost-effectiveness of the segmentation workflow. Our innovative framework addresses the limitations of current methods by combining a coarse segmentation network, a rejection network, and an interactive deep network known as Segment Anything Model (SAM). The coarse segmentation network automatically generates initial segmentation results, which are evaluated by the rejection network to estimate their quality. Low-quality results are flagged for user interaction, with the user providing a region of interest (ROI) enclosing the lesions, whereas for high-quality results, ROIs were cropped from the automatic segmentation. Both manually and automatically defined ROIs are fed into SAM to produce the final fine segmentation. This approach significantly reduces the annotation burden and achieves substantial improvements by flagging approximately 20% of the images with the lowest quality scores for manual annotation. With only half of the images manually annotated, the final segmentation accuracy is statistically indistinguishable from that achieved using full manual annotation. Although this paper focuses on prostate lesion segmentation from multimodality MRI, the framework can be adapted to other medical image segmentation applications to improve segmentation efficiency while maintaining high accuracy standards.

Choroidal vascular changes in early-stage myopic maculopathy from deep learning choroidal analysis: a hospital-based SS-OCT study

BackgroundThe objective of this study is to illustrate the changes in the choroidal vasculature in individuals with diffuse chorioretinal atrophy (DCA, early-stage myopic maculopathy) and investigate the association between them.MethodsThis study included 1418 highly myopic eyes from 720 participants aged 18 − 60 years from the Wenzhou High Myopia Cohort Study. These participants underwent comprehensive ophthalmic assessments. Myopic maculopathy classification followed the Meta-PM system, with pathological myopia defined as myopic maculopathy of DCA or severer. Eyes with myopic maculopathy categorized as no macular lesions (C0), tessellated fundus (C1), and DCA (C2) were enrolled in the analysis. Choroidal images were obtained from swept-source optical coherence tomography (SS-OCT), and the images were processed with a deep learning-based automatic segmentation algorithm and the Niblack auto-local threshold algorithm.ResultsDCA was detected in 247 eyes (17.4%). In comparison to eyes with C0, those with C2 exhibited significant reductions in choroidal thickness (ChT), luminal area (LA), and stromal area (SA) across all evaluated regions (all P < 0.001). An increase in choroidal vascular index (CVI) was observed in all regions, except for the nasal perifoveal (N2) and inferior perifoveal (I2) regions (all P < 0.01). Multivariable logistic regression analysis revealed a negative association between the presence of DCA and increases in choroidal LA and SA (odds ratio ≤ 0.099, P < 0.001). Multivariable linear regression analysis showed that the mean deviation of the visual field test was positively associated with LA and SA at the vertical meridian (B = 1.512, P < 0.001 for LA; B = 1.956, P < 0.001 for SA). Furthermore, the receiver operating characteristic curve analyses showed the optimal ChT to diagnose pathological myopia was 82.4 µm in the N2 region, the LA was 0.076 mm2 and the SA was 0.049 mm2, with area under the curves of 0.916, 0.908, and 0.895, respectively.ConclusionsThe results of this study indicated that both the presence of DCA and visual function impairment were associated with reductions in choroidal perfusion and stromal components. Moreover, we established threshold values for choroidal parameters in diagnosing pathological myopia, offering valuable references for clinical diagnosis and management.

Automatic Segmentation of Mediastinal Lymph Nodes and Blood Vessels in Endobronchial Ultrasound (EBUS) Images Using Deep Learning.

Endobronchial ultrasound (EBUS) is used in the minimally invasive sampling of thoracic lymph nodes. In lung cancer staging, the accurate assessment of mediastinal structures is essential but challenged by variations in anatomy, image quality, and operator-dependent image interpretation. This study aimed to automatically detect and segment mediastinal lymph nodes and blood vessels employing a novel U-Net architecture-based approach in EBUS images. A total of 1161 EBUS images from 40 patients were annotated. For training and validation, 882 images from 30 patients and 145 images from 5 patients were utilized. A separate set of 134 images was reserved for testing. For lymph node and blood vessel segmentation, the mean ± standard deviation (SD) values of the Dice similarity coefficient were 0.71 ± 0.35 and 0.76 ± 0.38, those of the precision were 0.69 ± 0.36 and 0.82 ± 0.22, those of the sensitivity were 0.71 ± 0.38 and 0.80 ± 0.25, those of the specificity were 0.98 ± 0.02 and 0.99 ± 0.01, and those of the F1 score were 0.85 ± 0.16 and 0.81 ± 0.21, respectively. The average processing and segmentation run-time per image was 55 ± 1 ms (mean ± SD). The new U-Net architecture-based approach (EBUS-AI) could automatically detect and segment mediastinal lymph nodes and blood vessels in EBUS images. The method performed well and was feasible and fast, enabling real-time automatic labeling.

Improving Access to Intracranial Volume Measurements for Clinical Use: Open-Source Method using Computed Tomography

Introduction: Three-dimensional measurements of intracranial volume (ICV) can guide clinical management of brain and skull disorders. However, widespread clinical access is limited by the scarcity of software methods for analyzing CT scans, which are more available than MRI images, and the inaccessibility of proprietary and expensive commercial software. The presented method can calculate ICV from CT scans, using an open-source software, 3D Slicer. Methods: The open-source workflow was optimized with a data-driven approach to find the optimal parameters for ICV accuracy. The accuracy of the open-source method was determined by comparing it to commercial and proprietary software with CT scans of pediatric hydrocephalic macrocephaly patients with craniosynostosis undergoing total vault reconstruction (N = 5 patients,15 scans). Results: An open-source pipeline that combines an initial semi-automatic segmentation of a coronal CT reconstruction with a fully automatic segmentation minimizes the ICV error. The open-source method shows excellent agreement with both the commercial and proprietary software methods (R2 = 0.998 and 95% confidence interval of best-fit line slope: [0.986; 1.047], [0.985;1.066] respectively). The mean percent difference of ICV measurements of the open-source software from the commercial software was −0.56% [95% CI: −1.08%, −0.028%] and from the proprietary software was −0.07% [95% CI: −1.26%, 1.1%]. The mean percent difference of ICV measurements of the commercial software from the proprietary software was 0.36% [95% Confidence Interval: −0.61%, 1.32%]. Conclusions: This is the first study comparing an open-source method for measuring ICV with commercial and proprietary options. A high degree of fidelity was found, confirming this open-source method as a viable option for clinicians who are looking to incorporate ICV measurements into their practice.