Ground Truth Segmentation Research Articles

Clinical capability of modern brain tumor segmentation models.

State-of-the-art automated segmentation methods achieve exceptionally high performance on the Brain Tumor Segmentation (BraTS) challenge, a dataset of uniformly processed and standardized magnetic resonance generated images (MRIs) of gliomas. However, a reasonable concern is that these models may not fare well on clinical MRIs that do not belong to the specially curated BraTS dataset. Research using the previous generation of deep learning models indicates significant performance loss on cross-institutional predictions. Here, we evaluate the cross-institutional applicability and generalzsability of state-of-the-art deep learning models on new clinicaldata. We train a state-of-the-art 3D U-Net model on the conventional BraTS dataset comprising low- and high-grade gliomas. We then evaluate the performance of this model for automatic tumor segmentation of brain tumors on in-house clinical data. This dataset contains MRIs of different tumor types, resolutions, and standardization than those found in the BraTS dataset. Ground truth segmentations to validate the automated segmentation for in-house clinical data were obtained from expert radiationoncologists. We report average Dice scores of 0.764, 0.648, and 0.61 for the whole tumor, tumor core, and enhancing tumor, respectively, in the clinical MRIs. These means are higher than numbers reported previously on same institution and cross-institution datasets of different origin using different methods. There is no statistically significant difference when comparing the dice scores to the inter-annotation variability between two expert clinical radiation oncologists. Although performance on the clinical data is lower than on the BraTS data, these numbers indicate that models trained on the BraTS dataset have impressive segmentation performance on previously unseen images obtained at a separate clinical institution. These images differ in the imaging resolutions, standardization pipelines, and tumor types from the BraTSdata. State-of-the-art deep learning models demonstrate promising performance on cross-institutional predictions. They considerably improve on previous models and can transfer knowledge to new types of brain tumors without additionalmodeling.

Image-Based Corrosion Detection in Ancillary Structures

Ancillary structures are essential for highways’ safe operationality but are mainly prone to environmental corrosion. The traditional way of inspecting ancillary structures is manned inspection, which is laborious, time-consuming, and unsafe for inspectors. In this paper, a novel image processing technique was developed for autonomous corrosion detection of in-service ancillary structures. The authors successfully leveraged corrosion features in the YCbCr color space as an alternative to the conventional red–green–blue (RGB) color space. The proposed method included a preprocessing operation including contrast adjustment, histogram equalization, adaptive histogram equalization, and optimum value determination of brightness. The effect of preprocessing was evaluated against a semantically segmented ground truth as a set of pixel-level annotated images. The false detection rate was higher in Otsu than in the global threshold method; therefore, the preprocessed images were converted to binary using the global threshold value. Finally, an average accuracy and true positive rate of 90% and 70%, respectively, were achieved for corrosion prediction in the YCbCr color space.

AxoNet 2.0: A Deep Learning-Based Tool for Morphometric Analysis of Retinal Ganglion Cell Axons.

Assessment of glaucomatous damage in animal models is facilitated by rapid and accurate quantification of retinal ganglion cell (RGC) axonal loss and morphologic change. However, manual assessment is extremely time- and labor-intensive. Here, we developed AxoNet 2.0, an automated deep learning (DL) tool that (i) counts normal-appearing RGC axons and (ii) quantifies their morphometry from light micrographs. A DL algorithm was trained to segment the axoplasm and myelin sheath of normal-appearing axons using manually-annotated rat optic nerve (ON) cross-sectional micrographs. Performance was quantified by various metrics (e.g., soft-Dice coefficient between predicted and ground-truth segmentations). We also quantified axon counts, axon density, and axon size distributions between hypertensive and control eyes and compared to literature reports. AxoNet 2.0 performed very well when compared to manual annotations of rat ON (R2 = 0.92 for automated vs. manual counts, soft-Dice coefficient = 0.81 ± 0.02, mean absolute percentage error in axonal morphometric outcomes < 15%). AxoNet 2.0 also showed promise for generalization, performing well on other animal models (R2 = 0.97 between automated versus manual counts for mice and 0.98 for non-human primates). As expected, the algorithm detected decreased in axon density in hypertensive rat eyes (P ≪ 0.001) with preferential loss of large axons (P < 0.001). AxoNet 2.0 provides a fast and nonsubjective tool to quantify both RGC axon counts and morphological features, thus assisting with assessing axonal damage in animal models of glaucomatous optic neuropathy. This deep learning approach will increase rigor of basic science studies designed to investigate RGC axon protection and regeneration.

A Self-Configuring Deep Learning Network for Segmentation of Temporal Bone Anatomy in Cone-Beam CT Imaging.

Preoperative planning for otologic or neurotologic procedures often requires manual segmentation of relevant structures, which can be tedious and time-consuming. Automated methods for segmenting multiple geometrically complex structures can not only streamline preoperative planning but also augment minimally invasive and/or robot-assisted procedures in this space. This study evaluates a state-of-the-art deep learning pipeline for semantic segmentation of temporal bone anatomy. A descriptive study of a segmentation network. Academic institution. A total of 15 high-resolution cone-beam temporal bone computed tomography (CT) data sets were included in this study. All images were co-registered, with relevant anatomical structures (eg, ossicles, inner ear, facial nerve, chorda tympani, bony labyrinth) manually segmented. Predicted segmentations from no new U-Net (nnU-Net), an open-source 3-dimensional semantic segmentation neural network, were compared against ground-truth segmentations using modified Hausdorff distances (mHD) and Dice scores. Fivefold cross-validation with nnU-Net between predicted and ground-truth labels were as follows: malleus (mHD: 0.044 ± 0.024 mm, dice: 0.914 ± 0.035), incus (mHD: 0.051 ± 0.027 mm, dice: 0.916 ± 0.034), stapes (mHD: 0.147 ± 0.113 mm, dice: 0.560 ± 0.106), bony labyrinth (mHD: 0.038 ± 0.031 mm, dice: 0.952 ± 0.017), and facial nerve (mHD: 0.139 ± 0.072 mm, dice: 0.862 ± 0.039). Comparison against atlas-based segmentation propagation showed significantly higher Dice scores for all structures (p < .05). Using an open-source deep learning pipeline, we demonstrate consistently submillimeter accuracy for semantic CT segmentation of temporal bone anatomy compared to hand-segmented labels. This pipeline has the potential to greatly improve preoperative planning workflows for a variety of otologic and neurotologic procedures and augment existing image guidance and robot-assisted systems for the temporal bone.

Cross-model consensus of explanations and beyond for image classification models: an empirical study

Existing interpretation algorithms have found that, even deep models make the same and right predictions on the same image, they might rely on different sets of input features for classification. However, among these sets of features, some common features might be used by the majority of models. In this paper, we are wondering what are the common features used by various models for classification and whether the models with better performance may favor those common features. For this purpose, our works uses an interpretation algorithm to attribute the importance of features (e.g., pixels or superpixels) as explanations, and proposes the cross-model consensus of explanations to capture the common features. Specifically, we first prepare a set of deep models as a committee, then deduce the explanation for every model, and obtain the consensus of explanations across the entire committee through voting. With the cross-model consensus of explanations, we conduct extensive experiments using 80+ models on 5 datasets/tasks. We find three interesting phenomena as follows: (1) the consensus obtained from image classification models is aligned with the ground truth of semantic segmentation; (2) we measure the similarity of the explanation result of each model in the committee to the consensus (namely consensus score), and find positive correlations between the consensus score and model performance; and (3) the consensus score coincidentally correlates to the interpretability.

Cattle Weight Estimation Using Fully and Weakly Supervised Segmentation from 2D Images

Weight information is important in cattle breeding because it can measure animal growth and be used to calculate the appropriate amount of daily feed. To estimate the weight, we developed an image-based method that does not stress cattle and requires no manual labor. From a 2D image, a mask was obtained by segmenting the animal and background, and weights were estimated using a deep neural network with residual connections by extracting weight-related features from the segmentation mask. Two image segmentation methods, fully and weakly supervised segmentation, were compared. The fully supervised segmentation method uses a Mask R-CNN model that learns the ground truth mask generated by labeling as the correct answer. The weakly supervised segmentation method uses an activation visualization map that is proposed in this study. The first method creates a more precise mask, but the second method does not require ground truth segmentation labeling. The body weight was estimated using statistical features of the segmented region. In experiments, the following performance results were obtained: a mean average error of 17.31 kg and mean absolute percentage error of 5.52% for fully supervised segmentation, and a mean average error of 35.91 kg and mean absolute percentage error of 10.1% for the weakly supervised segmentation.

Effects of Image Size on Deep Learning

In this work, the best size for late gadolinium enhancement (LGE) magnetic resonance imaging (MRI) images in the training dataset was determined to optimize deep learning training outcomes. Non-extra pixel and extra pixel interpolation algorithms were used to determine the new size of the LGE-MRI images. A novel strategy was introduced to handle interpolation masks and remove extra class labels in interpolated ground truth (GT) segmentation masks. The expectation maximization, weighted intensity, a priori information (EWA) algorithm was used for the quantification of myocardial infarction (MI) in automatically segmented LGE-MRI images. Arbitrary threshold, comparison of the sums, and sums of differences are methods used to estimate the relationship between semi-automatic or manual and fully automated quantification of myocardial infarction (MI) results. The relationship between semi-automatic and fully automated quantification of MI results was found to be closer in the case of bigger LGE MRI images (55.5% closer to manual results) than in the case of smaller LGE MRI images (22.2% closer to manual results).

Comparison of automated segmentation techniques for magnetic resonance images of the prostate

BackgroundContouring of anatomical regions is a crucial step in the medical workflow and is both time-consuming and prone to intra- and inter-observer variability. This study compares different strategies for automatic segmentation of the prostate in T2-weighted MRIs.MethodsThis study included 100 patients diagnosed with prostate adenocarcinoma who had undergone multi-parametric MRI and prostatectomy. From the T2-weighted MR images, ground truth segmentation masks were established by consensus from two expert radiologists. The prostate was then automatically contoured with six different methods: (1) a multi-atlas algorithm, (2) a proprietary algorithm in the Syngo.Via medical imaging software, and four deep learning models: (3) a V-net trained from scratch, (4) a pre-trained 2D U-net, (5) a GAN extension of the 2D U-net, and (6) a segmentation-adapted EfficientDet architecture. The resulting segmentations were compared and scored against the ground truth masks with one 70/30 and one 50/50 train/test data split. We also analyzed the association between segmentation performance and clinical variables.ResultsThe best performing method was the adapted EfficientDet (model 6), achieving a mean Dice coefficient of 0.914, a mean absolute volume difference of 5.9%, a mean surface distance (MSD) of 1.93 pixels, and a mean 95th percentile Hausdorff distance of 3.77 pixels. The deep learning models were less prone to serious errors (0.854 minimum Dice and 4.02 maximum MSD), and no significant relationship was found between segmentation performance and clinical variables.ConclusionsDeep learning-based segmentation techniques can consistently achieve Dice coefficients of 0.9 or above with as few as 50 training patients, regardless of architectural archetype. The atlas-based and Syngo.via methods found in commercial clinical software performed significantly worse (0.855-0.887 Dice).

Chákṣu: A glaucoma specific fundus image database

We introduce Chákṣu–a retinal fundus image database for the evaluation of computer-assisted glaucoma prescreening techniques. The database contains 1345 color fundus images acquired using three brands of commercially available fundus cameras. Each image is provided with the outlines for the optic disc (OD) and optic cup (OC) using smooth closed contours and a decision of normal versus glaucomatous by five expert ophthalmologists. In addition, segmentation ground-truths of the OD and OC are provided by fusing the expert annotations using the mean, median, majority, and Simultaneous Truth and Performance Level Estimation (STAPLE) algorithm. The performance indices show that the ground-truth agreement with the experts is the best with STAPLE algorithm, followed by majority, median, and mean. The vertical, horizontal, and area cup-to-disc ratios are provided based on the expert annotations. Image-wise glaucoma decisions are also provided based on majority voting among the experts. Chákṣu is the largest Indian-ethnicity-specific fundus image database with expert annotations and would aid in the development of artificial intelligence based glaucoma diagnostics.

Abstract TMP69: Random Rater Sampling For Deep Learning Algorithm To Segment Acute Ischemic Stroke On Non-contrast Computed Tomography

Introduction: The delineation of volume and location of acute ischemic brain tissue (AIBT) on Non-Contrast CT (NCCT) increases the efficiency of endovascular treatment decisions. However, the manual segmentation of the AIBT on NCCT is a challenging task and suffers from low inter-expert agreement. Whether supervised deep convolutional neural networks (CNN) are more accurate than expert raters remain to be determined, and the optimal ground truth segmentation of AIBT is unclear given the low inter-expert agreement. We hypothesized that randomly sampling ground truth segmentations of expert raters would enable the CNN to better approximate an accurate ground truth and increase performance. Methods: The data set consisted of 200 NCCT images (Figure 1a)) of acute ischemic stroke patients presenting within 6-16h and consenting to the DEFUSE3 trial. Three experienced neuroradiologists manually segmented the AIBT (Figure 1b)1.-3.). The aggregated validation sets of 5-fold-cross-validation were used to compare the (i) average inter-expert agreement, the average performance of (ii) three individual CNNs trained on each rater and (iii) one CNN trained with random rater sampling. Results: (iii) Random rater sampling (Figure 1c)) lead to a CNN performance superior to (ii) the average performance of individually trained CNNs. The reliability of a CNN goes beyond the human inter-expert agreement (i) (Surface Dice at Tolerance 5mm: 0.88 versus 0.68 and 0.67, respectively; Table 1). Conclusions: The volume and location of AIBT on NCCT can reliably be segmented by a CNN. The agreement between a randomly chosen rater and the CNN predictions surpasses the inter-expert agreement (Table 1).

Video Object Segmentation Using Kernelized Memory Network With Multiple Kernels.

Semi-supervised video object segmentation (VOS) is to predict the segment of a target object in a video when a ground truth segmentation mask for the target is given in the first frame. Recently, space-time memory networks (STM) have received significant attention as a promising approach for semi-supervised VOS. However, an important point has been overlooked in applying STM to VOS: The solution (=STM) is non-local, but the problem (=VOS) is predominantly local. To solve this mismatch between STM and VOS, we propose new VOS networks called kernelized memory network (KMN) and KMN with multiple kernels (KMN M). Our networks conduct not only Query-to-Memory matching but also Memory-to-Query matching. In Memory-to-Query matching, a kernel is employed to reduce the degree of non-localness of the STM. In addition, we present a Hide-and-Seek strategy in pre-training to handle occlusions effectively. The proposed networks surpass the state-of-the-art results on standard benchmarks by a significant margin (+4% in JM on DAVIS 2017 test-dev set). The runtimes of our proposed KMN and KMN M on DAVIS 2016 validation set are 0.12 and 0.13 seconds per frame, respectively, and the two networks have similar computation times to STM.

A crowdsourced dataset of aerial images with annotated solar photovoltaic arrays and installation metadata

Photovoltaic (PV) energy generation plays a crucial role in the energy transition. Small-scale, rooftop PV installations are deployed at an unprecedented pace, and their safe integration into the grid requires up-to-date, high-quality information. Overhead imagery is increasingly being used to improve the knowledge of rooftop PV installations with machine learning models capable of automatically mapping these installations. However, these models cannot be reliably transferred from one region or imagery source to another without incurring a decrease in accuracy. To address this issue, known as distribution shift, and foster the development of PV array mapping pipelines, we propose a dataset containing aerial images, segmentation masks, and installation metadata (i.e., technical characteristics). We provide installation metadata for more than 28000 installations. We supply ground truth segmentation masks for 13000 installations, including 7000 with annotations for two different image providers. Finally, we provide installation metadata that matches the annotation for more than 8000 installations. Dataset applications include end-to-end PV registry construction, robust PV installations mapping, and analysis of crowdsourced datasets.

A Deep Learning Method for Quantification of Femoral Head Necrosis Based on Routine Hip MRI for Improved Surgical Decision Making.

(1) Background: To evaluate the performance of a deep learning model to automatically segment femoral head necrosis (FHN) based on a standard 2D MRI sequence compared to manual segmentations for 3D quantification of FHN. (2) Methods: Twenty-six patients (thirty hips) with avascular necrosis underwent preoperative MR arthrography including a coronal 2D PD-w sequence and a 3D T1 VIBE sequence. Manual ground truth segmentations of the necrotic and unaffected bone were then performed by an expert reader to train a self-configuring nnU-Net model. Testing of the network performance was performed using a 5-fold cross-validation and Dice coefficients were calculated. In addition, performance across the three segmentations were compared using six parameters: volume of necrosis, volume of unaffected bone, percent of necrotic bone volume, surface of necrotic bone, unaffected femoral head surface, and percent of necrotic femoral head surface area. (3) Results: Comparison between the manual 3D and manual 2D segmentations as well as 2D with the automatic model yielded significant, strong correlations (Rp > 0.9) across all six parameters of necrosis. Dice coefficients between manual- and automated 2D segmentations of necrotic- and unaffected bone were 75 ± 15% and 91 ± 5%, respectively. None of the six parameters of FHN differed between the manual and automated 2D segmentations and showed strong correlations (Rp > 0.9). Necrotic volume and surface area showed significant differences (all p < 0.05) between early and advanced ARCO grading as opposed to the modified Kerboul angle, which was comparable between both groups (p > 0.05). (4) Conclusions: Our deep learning model to automatically segment femoral necrosis based on a routine hip MRI was highly accurate. Coupled with improved quantification for volume and surface area, as opposed to 2D angles, staging and course of treatment can become better tailored to patients with varying degrees of AVN.

Multilesion Segmentations in Patients with Intracerebral Hemorrhage: Reliability of ICH, IVH and PHE Masks.

Background and Purpose: Fully automated methods for segmentation and volume quantification of intraparenchymal hemorrhage (ICH), intraventricular hemorrhage extension (IVH), and perihematomal edema (PHE) are gaining increasing interest. Yet, reliabilities demonstrate considerable variances amongst each other. Our aim was therefore to evaluate both the intra- and interrater reliability of ICH, IVH and PHE on ground-truth segmentation masks. Methods: Patients with primary spontaneous ICH were retrospectively included from a German tertiary stroke center (Charité Berlin; January 2016−June 2020). Baseline and follow-up non-contrast Computed Tomography (NCCT) scans were analyzed for ICH, IVH, and PHE volume quantification by two radiology residents. Raters were blinded to all demographic and outcome data. Inter- and intrarater agreements were determined by calculating the Intraclass Correlation Coefficient (ICC) for a randomly selected set of patients with ICH, IVH, and PHE. Results: 100 out of 670 patients were included in the analysis. Interrater agreements ranged from an ICC of 0.998 for ICH (95% CI [0.993; 0.997]), to an ICC of 0.979 for IVH (95% CI [0.984; 0.993]), and an ICC of 0.886 for PHE (95% CI [0.760; 0.938]), all p-values < 0.001. Intrarater agreements ranged from an ICC of 0.997 for ICH (95% CI [0.996; 0.998]), to an ICC of 0.995 for IVH (95% CI [0.992; 0.996]), and an ICC of 0.980 for PHE (95% CI [0.971; 0.987]), all p-values < 0.001. Conclusion Manual segmentations of ICH, IVH, and PHE demonstrate good-to-excellent inter- and intrarater reliabilities, with the highest agreement for ICH and IVH and lowest for PHE. Therefore, the degree of variances reported in fully automated quantification methods might be related amongst others to variances in ground-truth masks.

MITEA: A dataset for machine learning segmentation of the left ventricle in 3D echocardiography using subject-specific labels from cardiac magnetic resonance imaging.

Segmentation of the left ventricle (LV) in echocardiography is an important task for the quantification of volume and mass in heart disease. Continuing advances in echocardiography have extended imaging capabilities into the 3D domain, subsequently overcoming the geometric assumptions associated with conventional 2D acquisitions. Nevertheless, the analysis of 3D echocardiography (3DE) poses several challenges associated with limited spatial resolution, poor contrast-to-noise ratio, complex noise characteristics, and image anisotropy. To develop automated methods for 3DE analysis, a sufficiently large, labeled dataset is typically required. However, ground truth segmentations have historically been difficult to obtain due to the high inter-observer variability associated with manual analysis. We address this lack of expert consensus by registering labels derived from higher-resolution subject-specific cardiac magnetic resonance (CMR) images, producing 536 annotated 3DE images from 143 human subjects (10 of which were excluded). This heterogeneous population consists of healthy controls and patients with cardiac disease, across a range of demographics. To demonstrate the utility of such a dataset, a state-of-the-art, self-configuring deep learning network for semantic segmentation was employed for automated 3DE analysis. Using the proposed dataset for training, the network produced measurement biases of -9 ± 16 ml, -1 ± 10 ml, -2 ± 5 %, and 5 ± 23 g, for end-diastolic volume, end-systolic volume, ejection fraction, and mass, respectively, outperforming an expert human observer in terms of accuracy as well as scan-rescan reproducibility. As part of the Cardiac Atlas Project, we present here a large, publicly available 3DE dataset with ground truth labels that leverage the higher resolution and contrast of CMR, to provide a new benchmark for automated 3DE analysis. Such an approach not only reduces the effect of observer-specific bias present in manual 3DE annotations, but also enables the development of analysis techniques which exhibit better agreement with CMR compared to conventional methods. This represents an important step for enabling more efficient and accurate diagnostic and prognostic information to be obtained from echocardiography.

Language-Guided Multi-Granularity Context Aggregation for Temporal Sentence Grounding

Temporal sentence grounding in videos is a crucial task in vision-language learning. Its goal is retrieving a video segment from an untrimmed video that semantically corresponds to a natural language query. A video usually contains multiple semantic events, which are rarely isolated. They tend to be temporally ordered and semantically correlated ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e.g.</i> , some event is often the precursor of another event). To precisely localize a semantic moment from a video, it is critical to effectively extract and aggregate multi-granularity contextual information, including the fine-grained local context around the moment-related video segment (in short snippet-level) and coarse-grained semantic correlation (in segment-level). Additionally, a second main insight in this work is that the above context aggregation should be favorably guided by the queries, rather than fully query-agnostic. Putting above ideas together, we here present a new network that does language-guided multi-granularity context aggregation. It is comprised of two major modules. The core of the first module is a novel language-guided temporal adaptive convolution (LTAC) devised to extract fine-grained information over video snippets around the ground-truth video segment. It decomposes a convolution into two channel-oriented / temporal-oriented ones. In particular, the convolutional channels are supposed to be more susceptible to queries, thus we learn to generate a dynamic channel-oriented kernel with respect to the querying sentence. As a second module, we propose a language-guided global relation block (LGRB) that extracts video-level context. It augments the contextual feature by using a multi-scale temporal attention that tackles the scale variation of ground-truth video segments, and a multi-modal semantic attention that relies on syntactic of the query. For the validation purpose, we have conducted comprehensive experiments on two popularly-adopted video benchmarks ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i> , ActivityNet Captions and Charades-STA). All experimental results and ablation studies have clearly corroborated the effectiveness of our model designs, outstripping prior state-of-the-art methods in terms of major performance metrics for the task.

Glass Segmentation with RGB-Thermal Image Pairs.

This paper proposes a new glass segmentation method utilizing paired RGB and thermal images. Due to the large difference between the transmission property of visible light and that of the thermal energy through the glass where most glass is transparent to the visible light but opaque to thermal energy, glass regions of a scene are made more distinguishable with a pair of RGB and thermal images than solely with an RGB image. To exploit such a unique property, we propose a neural network architecture that effectively combines an RGB-thermal image pair with a new multi-modal fusion module based on attention, and integrate CNN and transformer to extract local features and non-local dependencies, respectively. As well, we have collected a new dataset containing 5551 RGB-thermal image pairs with ground-truth segmentation annotations. The qualitative and quantitative evaluations demonstrate the effectiveness of the proposed approach on fusing RGB and thermal data for glass segmentation. Our code and data are available at https://github.com/Dong-Huo/RGB-T-Glass-Segmentation.

Quality assurance assessment of intra-acquisition diffusion-weighted and T2-weighted magnetic resonance imaging registration and contour propagation for head and neck cancer radiotherapy.

Adequate image registration of anatomical and functional magnetic resonance imaging (MRI) scans is necessary for MR-guided head and neck cancer (HNC) adaptive radiotherapy planning. Despite the quantitative capabilities of diffusion-weighted imaging (DWI) MRI for treatment plan adaptation, geometric distortion remains a considerable limitation. Therefore, we systematically investigated various deformable image registration (DIR) methods to co-register DWI and T2-weighted (T2W) images. We compared three commercial (ADMIRE, Velocity, Raystation) and three open-source (Elastix with default settings [Elastix Default], Elastix with parameter set 23 [Elastix 23], Demons) post-acquisition DIR methods applied to T2W and DWI MRI images acquired during the same imaging session in twenty immobilized HNC patients. In addition, we used the non-registered images (None) as a control comparator. Ground-truth segmentations of radiotherapy structures (tumour and organs at risk) were generated by a physician expert on both image sequences. For each registration approach, structures were propagated from T2W to DWI images. These propagated structures were then compared with ground-truth DWI structures using the Dice similarity coefficient and mean surface distance. 19 left submandibular glands, 18 right submandibular glands, 20left parotid glands, 20 right parotid glands, 20 spinal cords, and 12 tumours were delineated. Most DIR methods took <30s to execute per case, with the exception of Elastix 23 which took ∼458s to execute per case. ADMIRE and Elastix 23 demonstrated improved performance over None for all metrics and structures (Bonferroni-corrected p<0.05), while the other methods did not. Moreover, ADMIRE and Elastix 23 significantly improved performance in individual and pooled analysis compared to all other methods. The ADMIRE DIR method offers improved geometric performance with reasonable execution time so should be favoured for registering T2W and DWI images acquired during the same scan session in HNC patients. These results are important to ensure the appropriate selection of registration strategies for MR-guided radiotherapy.

Combined Detection and Segmentation of Overlapping Erythrocytes in Microscopy Images Using Morphological Image Processing

Segmentation of clusters of erythrocytes into their constituent single cells is a procedure needed in various biomedical applications related to microscopy images. This task is part of the general problem of splitting clumps of objects in images which continues being an open research topic in the Image Processing field. This work presents a unified morphological method to detect and segment clusters of erythrocytes in microscopy images, and proposes two main contributions. The first one is to formulate and evaluate a method to detect clusters as connected components in binary images, obtained from a previous coarse segmentation, which is not capable of further dividing a cluster into its constituent cells. Secondly, to propose the best alternative to split the clusters into their constituent individual cells after evaluating three algorithms based in the combination of the transforms: H-maxima, weighted external distance and marker-controlled watershed. Evaluation of the proposed cluster detection methods was made in terms of standard measures of effectiveness. Segmentation accuracy was evaluated comparing the segmented objects obtained to a manually segmented ground truth, by means of the Jaccard index. Then the Friedman test allowed validating the advantages of the proposed method in comparison to the other alternatives studied here.

Automatic lesion detection and segmentation in 18F-flutemetamol positron emission tomography images using deep learning

BackgroundBeta amyloid in the brain, which was originally confirmed by post-mortem examinations, can now be confirmed in living patients using amyloid positron emission tomography (PET) tracers, and the accuracy of diagnosis can be improved by beta amyloid plaque confirmation in patients. Amyloid deposition in the brain is often associated with the expression of dementia. Hence, it is important to identify the anatomically and functionally meaningful areas of the human brain cortex surface using PET to diagnose the possibility of developing dementia. In this study, we demonstrated the validity of automated 18F-flutemetamol PET lesion detection and segmentation based on a complete 2D U-Net convolutional neural network via masking treatment strategies.MethodsPET data were first normalized by volume and divided into five amyloid accumulation zones through axial, coronary, and thalamic slices. A single U-Net was trained using a divided dataset for one of these zones. Ground truth segmentations were obtained by manual delineation and thresholding (1.5 × background).ResultsThe following intersection over union values were obtained for the various slices in the verification dataset: frontal lobe axial/sagittal: 0.733/0.804; posterior cingulate cortex and precuneus coronal/sagittal: 0.661/0.726; lateral temporal lobe axial/coronal: 0.864/0.892; parietal lobe axial/coronal: 0.542/0.759; and striatum axial/sagittal: 0.679/0.752. The U-Net convolutional neural network architecture allowed fully automated 2D division of the 18F-flutemetamol PET brain images of Alzheimer's patients.ConclusionsAs dementia should be tested and evaluated in various ways, there is a need for artificial intelligence programs. This study can serve as a reference for future studies using auxiliary roles and research in Alzheimer's diagnosis.