Automatic Segmentation Technique Research Articles

Uncertainty quantification in multi-class segmentation: Comparison between Bayesian and non-Bayesian approaches in a clinical perspective.

Automatic segmentation techniques based on Convolutional Neural Networks (CNNs) are widely adopted to automatically identify any structure of interest from a medical image, as they are not time consuming and not subject to high intra- and inter-operator variability. However, the adoption of these approaches in clinical practice is slowed down by some factors, such as the difficulty in providing an accurate quantification of their uncertainty. This work aims to evaluate the uncertainty quantification provided by two Bayesian and two non-Bayesian approaches for a multi-class segmentation problem, and to compare the risk propensity among these approaches, considering CT images of patients affected by renal cancer (RC). Four uncertainty quantification approaches were implemented in this work, based on a benchmark CNN currently employed in medical image segmentation: two Bayesian CNNs with different regularizations (Dropout and DropConnect), named BDR and BDC, an ensemble method (Ens) and a test-time augmentation (TTA) method. They were compared in terms of segmentation accuracy, using the Dice score, uncertainty quantification, using the ratio of correct-certain pixels (RCC) and incorrect-uncertain pixels (RIU), and with respect to inter-observer variability in manual segmentation. They were trained with the Kidney and Kidney Tumor Segmentation Challenge launched in 2021 (Kits21), for which multi-class segmentations of kidney, RC, and cyst on 300 CT volumes are available. Moreover, they were tested considering this and other two public renal CT datasets. Accuracy results achieved large differences across the structures of interest for all approaches, with an average Dice score of 0.92, 0.58, and 0.21 for kidney, tumor, and cyst, respectively. In terms of uncertainties, TTA provided the highest uncertainty, followed by Ens and BDC, whereas BDR provided the lowest, and minimized the number of incorrect certain pixels worse than the other approaches. Again, large differences were seen across the three structures in terms of RCC and RIU. These metrics were associated with different risk propensity, as BDR was the most risk-taking approach, able to provide higher accuracy in its prediction, but failing to assign uncertainty on incorrect segmentation in every case. The other three approaches were more conservative, providing large uncertainty regions, with the drawback of giving alert also on correct areas. Finally, the analysis of the inter-observer segmentation variability showed a significant variation among the four approaches on the external dataset, with BDR reporting the lowest agreement (Dice = 0.82), and TTA obtaining the highest score (Dice = 0.94). Our outcomes highlight the importance of quantifying the segmentation uncertainty and that decision-makers can choose the approach most in line with the risk propensity degree required by the application and theirpolicy.

DTONet a Lightweight Model for Melanoma Segmentation.

With the further development of neural networks, automatic segmentation techniques for melanoma are becoming increasingly mature, especially under the conditions of abundant hardware resources. This allows for the accuracy of segmentation to be improved by increasing the complexity and computational capacity of the model. However, a new problem arises when it comes to actual applications, as there may not be the high-end hardware available, especially in hospitals and among the general public, who may have limited computing resources. In response to this situation, this paper proposes a lightweight deep learning network that can achieve high segmentation accuracy with minimal resource consumption. We introduce a network called DTONet (double-tailed octave network), which was specifically designed for this purpose. Its computational parameter count is only 30,859, which is 1/256th of the mainstream UNet model. Despite its reduced complexity, DTONet demonstrates superior performance in terms of accuracy, with an IOU improvement over other similar models. To validate the generalization capability of this model, we conducted tests on the PH2 dataset, and the results still outperformed existing models. Therefore, the proposed DTONet network exhibits excellent generalization ability and is sufficiently outstanding.

Automatic Brain Tissue and Lesion Segmentation and Multi-Parametric Mapping of Contrast-Enhancing Gliomas without the Injection of Contrast Agents: A Preliminary Study.

This study aimed to develop a rapid, 1 mm3 isotropic resolution, whole-brain MRI technique for automatic lesion segmentation and multi-parametric mapping without using contrast by continuously applying balanced steady-state free precession with inversion pulses throughout incomplete inversion recovery in a single 6 min scan. Modified k-means clustering was performed for automatic brain tissue and lesion segmentation using distinct signal evolutions that contained mixed T1/T2/magnetization transfer properties. Multi-compartment modeling was used to derive quantitative multi-parametric maps for tissue characterization. Fourteen patients with contrast-enhancing gliomas were scanned with this sequence prior to the injection of a contrast agent, and their segmented lesions were compared to conventionally defined manual segmentations of T2-hyperintense and contrast-enhancing lesions. Simultaneous T1, T2, and macromolecular proton fraction maps were generated and compared to conventional 2D T1 and T2 mapping and myelination water fraction mapping acquired with MAGiC. The lesion volumes defined with the new method were comparable to the manual segmentations (r = 0.70, p < 0.01; t-test p > 0.05). The T1, T2, and macromolecular proton fraction mapping values of the whole brain were comparable to the reference values and could distinguish different brain tissues and lesion types (p < 0.05), including infiltrating tumor regions within the T2-lesion. Highly efficient, whole-brain, multi-contrast imaging facilitated automatic lesion segmentation and quantitative multi-parametric mapping without contrast, highlighting its potential value in the clinic when gadolinium is contraindicated.

See Elegans: Simple-to-use, accurate, and automatic 3D detection of neural activity from densely packed neurons.

In the emerging field of whole-brain imaging at single-cell resolution, which represents one of the new frontiers to investigate the link between brain activity and behavior, the nematode Caenorhabditis elegans offers one of the most characterized models for systems neuroscience. Whole-brain recordings consist of 3D time series of volumes that need to be processed to obtain neuronal traces. Current solutions for this task are either computationally demanding or limited to specific acquisition setups. Here, we propose See Elegans, a direct programming algorithm that combines different techniques for automatic neuron segmentation and tracking without the need for the RFP channel, and we compare it with other available algorithms. While outperforming them in most cases, our solution offers a novel method to guide the identification of a subset of head neurons based on position and activity. The built-in interface allows the user to follow and manually curate each of the processing steps. See Elegans is thus a simple-to-use interface aimed at speeding up the post-processing of volumetric calcium imaging recordings while maintaining a high level of accuracy and low computational demands. (Contact: enrico.lanza@iit.it).

Main motivations and barriers to pro-environmental behaviour: a study from the employee’s perspective

Purpose The purpose of this study is to identify groups of employees based on their motivations, detecting the main barriers that may influence their willingness to participate in the pro-environmental initiatives proposed by their employer. Design/methodology/approach To identify the different groups of employees, an online survey was conducted, and the Chi-square automatic interaction detection algorithm segmentation technique was used with a sample of 483 employees from 9 Latin American universities. Findings The results allowed us to identify various segments, in which the main obstacle linked to intrinsic motivation is the university culture and, to a lesser extent, the lack of equipment, while for extrinsic motivation, the lack of infrastructure is the most determining factor. Likewise, the results reflect that, compared to the less motivated employees, those who show greater motivation (both intrinsic and extrinsic) are the ones who encounter the greatest barriers, so that the perceptions of the most motivated, as expert observers, help to identify the main obstacles that organisations must remove to promote pro-environmental behaviours among staff members. Practical implications The results obtained help to guide the representatives or organisational leaders on the actions that generate the greatest impact in the mitigation of climate change from a motivational approach of behavioural prediction. Social implications This study contributes to a more sustainable society by developing an understanding of how employees react to issues related to climate change. Knowing the perceptions of employees can be a turning point so that other members of society can get involved in pro-environmental behaviours. Originality/value Many studies have analysed the intrinsic and extrinsic motivations of employees to engage in pro-environmental behaviours; however, as far as the authors are aware, this has not been analysed from the perspective of barriers to motivation.

Full virtual patient generated by artificial intelligence-driven integrated segmentation of craniomaxillofacial structures from CBCT images

ObjectivesTo assess the performance, time-efficiency, and consistency of a convolutional neural network (CNN) based automated approach for integrated segmentation of craniomaxillofacial structures compared with semi-automated method for creating a virtual patient using cone beam computed tomography (CBCT) scans. MethodsThirty CBCT scans were selected. Six craniomaxillofacial structures, encompassing the maxillofacial complex bones, maxillary sinus, dentition, mandible, mandibular canal, and pharyngeal airway space, were segmented on these scans using semi-automated and composite of previously validated CNN-based automated segmentation techniques for individual structures. A qualitative assessment of the automated segmentation revealed the need for minor refinements, which were manually corrected. These refined segmentations served as a reference for comparing semi-automated and automated integrated segmentations. ResultsThe majority of minor adjustments with the automated approach involved under-segmentation of sinus mucosal thickening and regions with reduced bone thickness within the maxillofacial complex. The automated and the semi-automated approaches required an average time of 1.1 min and 48.4 min, respectively. The automated method demonstrated a greater degree of similarity (99.6 %) to the reference than the semi-automated approach (88.3 %). The standard deviation values for all metrics with the automated approach were low, indicating a high consistency. ConclusionsThe CNN-driven integrated segmentation approach proved to be accurate, time-efficient, and consistent for creating a CBCT-derived virtual patient through simultaneous segmentation of craniomaxillofacial structures. Clinical relevanceThe creation of a virtual orofacial patient using an automated approach could potentially transform personalized digital workflows. This advancement could be particularly beneficial for treatment planning in a variety of dental and maxillofacial specialties.

Image Segmentation of Triple-Negative Breast Cancer by Incorporating Multiscale and Parallel Attention Mechanisms

Breast cancer is a highly prevalent cancer. Triple-negative breast cancer (TNBC) is more likely to recur and metastasize than other subtypes of breast cancer. Research on the treatment of TNBC is of great importance, and accurate segmentation of the breast lesion area is an important step in the treatment of TNBC. Currently, the gold standard for tumor segmentation is still sketched manually by doctors, which requires expertise in the field of medical imaging and consumes a great deal of doctors’ time and energy. Automatic segmentation of breast cancer not only reduces the burden of doctors but also improves work efficiency. Therefore, it is of great significance to study the automatic segmentation technique for breast cancer lesion regions. In this paper, a deep-learning-based automatic segmentation algorithm for TNBC images is proposed. The experimental data were dynamic contrast-enhanced magnetic resonance imaging TNBC dataset provided by the Cancer Hospital of Zhengzhou University. The experiments were analyzed by comparing several models with UNet, Attention-UNet, ResUNet, and SegNet and using evaluation indexes such as Dice score and Iou. Compared to UNet, Attention-UNet, ResUNet, and SegNet, the proposed method improved the Dice score by 2.1%, 1.54%, 0.88%, and 9.65%, respectively. The experimental results show that the proposed deep-learning-based TNBC image segmentation model can effectively improve the segmentation performance of TNBC tumors.

Validation of QyScore’s® fully automated quantitative image segmentation tools against expert manual gold‐standards

AbstractBackgroundQuantitative imaging provides valuable information for early detection, disease progression and treatment response monitoring. Manual segmentation is the current gold‐standard, but is prohibitively labour‐intensive for large scale use, such as in clinical routine, and suffers from individual variability. Thus, there is a substantial unmet need for validation of automatic segmentation techniques that perform as accurately as this labour‐intensive manual gold‐standard.MethodThe validation cohort consisted of 50 individuals with multiple diagnoses, wide age range (18‐86 yrs), balanced for sex and acquired on multiple scanners for improved generalizability of quantification results (Table 1). Three expert neuroradiologists each manually segmented the caudate, putamen, globus pallidus, thalamus, cerebellum, brainstem, and lateral ventricles on 3DT1 images using itk‐SNAP (http://www.itksnap.org/) software. A consensus expert segmentation was then derived using the STAPLE (Simultaneous Truth and Performance Level Estimation) algorithm (Wakefield et al., (2004)) for each region and compared with the fully automated segmentations produced using QyScore®, a CE‐marked and FDA‐cleared neuroimaging medical device. Performance was investigated using the Dice Similarity Coefficient (DSC) and concordance assessed with plotted linear regression.ResultMean, standard deviation and confidence intervals of the DSC demonstrated a high degree of agreement between the consensus manual gold‐standard and QyScore®’s automated segmentations (Table 2). Importantly, this agreement remained consistent following stratification by field strength, demonstrating generalizability to most clinical imaging centres. For 1.5T scanners the mean DSC was 0.81, 0.89, 0.79, 0.86, 0.94, 0.93 and 0.91 for the caudate, putamen, globus pallidus, thalamus, cerebellum, brainstem, and lateral ventricles respectively. For the 3T scanners, this was equivalent or marginally better at 0.84, 0.89, 0.82, 0.86, 0.94, 0.94, and 0.93 respectively. In addition, strong concordance was demonstrated between the volumes, expressed as a percentage of intracranial volume (%ICV), obtained by the automated medical device QyScore® and the consensus of the manual gold‐standard segmentation (Figure 1: Table 2).ConclusionQyScore® produces fast reliable automated segmentations with comparable accuracy to expert neuroradiologists. These findings support the implementation of QyScore® in clinical trials and in clinical routine to provide quantitative image analysis in support of diagnosis and monitoring of disease progression and treatment response.

Validation of QyScore’s® fully automated quantitative image segmentation tools against expert manual gold‐standards

AbstractBackgroundQuantitative imaging provides valuable information for early detection, disease progression and treatment response monitoring. Manual segmentation is the current gold‐standard, but is prohibitively labour‐intensive for large scale use, such as in clinical routine, and suffers from individual variability. Thus, there is a substantial unmet need for validation of automatic segmentation techniques that perform as accurately as this labour‐intensive manual gold‐standard.MethodThe validation cohort consisted of 50 individuals with multiple diagnoses, wide age range (18‐86 yrs), balanced for sex and acquired on multiple scanners for improved generalizability of quantification results (Table 1). Three expert neuroradiologists each manually segmented the caudate, putamen, globus pallidus, thalamus, cerebellum, brainstem, and lateral ventricles on 3DT1 images using itk‐SNAP (http://www.itksnap.org/) software. A consensus expert segmentation was then derived using the STAPLE (Simultaneous Truth and Performance Level Estimation) algorithm (Wakefield et al., (2004)) for each region and compared with the fully automated segmentations produced using QyScore®, a CE‐marked and FDA‐cleared neuroimaging medical device. Performance was investigated using the Dice Similarity Coefficient (DSC) and concordance assessed with plotted linear regression.ResultMean, standard deviation and confidence intervals of the DSC demonstrated a high degree of agreement between the consensus manual gold‐standard and QyScore®’s automated segmentations (Table 2). Importantly, this agreement remained consistent following stratification by field strength, demonstrating generalizability to most clinical imaging centres. For 1.5T scanners the mean DSC was 0.81, 0.89, 0.79, 0.86, 0.94, 0.93 and 0.91 for the caudate, putamen, globus pallidus, thalamus, cerebellum, brainstem, and lateral ventricles respectively. For the 3T scanners, this was equivalent or marginally better at 0.84, 0.89, 0.82, 0.86, 0.94, 0.94, and 0.93 respectively. In addition, strong concordance was demonstrated between the volumes, expressed as a percentage of intracranial volume (%ICV), obtained by the automated medical device QyScore® and the consensus of the manual gold‐standard segmentation (Figure 1: Table 2).ConclusionQyScore® produces fast reliable automated segmentations with comparable accuracy to expert neuroradiologists. These findings support the implementation of QyScore® in clinical trials and in clinical routine to provide quantitative image analysis in support of diagnosis and monitoring of disease progression and treatment response.

Complex-Phase Steel Microstructure Segmentation Using UNet: Analysis across Different Magnifications and Steel Types

The quantification of the phase fraction is critical in materials science, bridging the gap between material composition, processing techniques, microstructure, and resultant properties. Traditional methods involving manual annotation are precise but labor-intensive and prone to human inaccuracies. We propose an automated segmentation technique for high-tensile strength alloy steel, where the complexity of microstructures presents considerable challenges. Our method leverages the UNet architecture, originally developed for biomedical image segmentation, and optimizes its performance via careful hyper-parameter selection and data augmentation. We employ Electron Backscatter Diffraction (EBSD) imagery for complex-phase segmentation and utilize a combined loss function to capture both textural and structural characteristics of the microstructures. Additionally, this work is the first to examine the scalability of the model across varying magnifications and types of steel and achieves high accuracy in terms of dice scores demonstrating the adaptability and robustness of the model.

Robust U-Net-Based Approach for Accurate Brain Tumor Segmentation Using Multimodal MRI Data

Detecting and quantifying the extent of brain tumors poses a formidable challenge in medical centers. Magnetic Resonance Imaging (MRI) has developed as a non-invasive brain cancers' primary diagnostic tool, offering the crucial advantage of avoiding ionizing radiation. Brain tumor manually segmented boundaries within 3D MRI volumes is an exceedingly time-intensive task, heavily reliant on operator expertise. Among brain tumors, gliomas stand out as the prevalent and highly malignant, significantly impacting patients' life expectancy, particularly at their highest grade. Recognizing the pressing need for a reliable, completely automatic segmentation technique to efficiently assess tumor extent, this study introduces a robust approach. A completely automated brain tumor segmentation method is proposed, leveraging U-Net-based deep convolutional networks. This approach underwent rigorous evaluation on the Multimodal Brain Tumor Image Segmentation BraTS-19 dataset a widely recognized medical image analysis dataset featuring multimodal MRI scans of brain tumors, including glioblastoma, anaplastic astrocytoma, and lower-grade glioma, coupled with corresponding manual tumor segmentations. This dataset serves as a pivotal resource for advancing automatic brain tumor segmentation techniques and assessing their performance using metrics like the Dice score, which achieved 92% for entire tumor. Cross-validation results affirm the efficiency and promise of our method in achieving accurate segmentation.

Brain tumor feature extraction and edge enhancement algorithm based on U-Net network

BackgroundStatistics show that each year more than 100,000 patients pass away from brain tumors. Due to the diverse morphology, hazy boundaries, or unbalanced categories of medical data lesions, segmentation prediction of brain tumors has significant challenges. PurposeIn this thesis, we highlight EAV-UNet, a system designed to accurately detect lesion regions. Optimizing feature extraction, utilizing automatic segmentation techniques to detect anomalous regions, and strengthening the structure. We prioritize the segmentation problem of lesion regions, especially in cases where the margins of the tumor are more hazy. MethodsThe VGG-19 network structure is incorporated into the coding stage of the U-Net, resulting in a deeper network structure, and an attention mechanism module is introduced to augment the feature information. Additionally, an edge detection module is added to the encoder to extract edge information in the image, which is then passed to the decoder to aid in reconstructing the original image. Our method uses the VGG-19 in place of the U-Net encoder. To strengthen feature details, we integrate a CBAM (Channel and Spatial Attention Mechanism) module into the decoder to enhance it. To extract vital edge details from the data, we incorporate an edge recognition section into the encoder. ResultsAll evaluation metrics show major improvements with our recommended EAV-UNet technique, which is based on a thorough analysis of experimental data. Specifically, for low contrast and blurry lesion edge images, the EAV-Unet method consistently produces forecasts that are very similar to the initial images. This technique reduced the Hausdorff distance to 1.82, achieved an F1 score of 96.1%, and attained a precision of 93.2% on Dataset 1. It obtained an F1 score of 76.8%, a Precision of 85.3%, and a Hausdorff distance reduction to 1.31 on Dataset 2. Dataset 3 displayed a Hausdorff distance cut in 2.30, an F1 score of 86.9%, and Precision of 95.3%. ConclusionsWe conducted extensive segmentation experiments using various datasets related to brain tumors. We refined the network architecture by employing smaller convolutional kernels in our strategy. To further improve segmentation accuracy, we integrated attention modules and an edge enhancement module to reinforce edge information and boost attention scores.

Organ Specific Deep Learning-Based Correction of Inaccurate Auto-Segmentation on Abdominal MRI

Deep learning-based automatic segmentation (DLAS) techniques offer limited success for abdominal organs on MRI, requiring substantial editing time. We have previously developed a deep learning based automatic contour correction (ACC) technique that can correct for inaccurate DLAS contours of bowels on MRI, reducing the manual editing time in MR-guided online adaptive radiation therapy (MRgART). This study aims to develop deep learning-based ACC models for pancreas and duodenum that are particularly difficult to contour either manually or with DLAS. Dense UNet, a deep learning algorithm that combines UNet with dense blocks, was trained to create ACC models. Organ-specific models were trained for pancreas and duodenum contours obtained from a research DLAS tool on MRIs from a 1.5T MR-Linac. The training dataset contained MRI slices paired with DLAS contours from 54 abdominal MRL sets along with ground truth contours and 540 additional augmented sets created by shifting, rotating, and scaling each organ along with the contours and varying the noise and bias field for each patient set. Each DLAS contour was classified into the acceptable (no additional edits required), minor edit (only simple edits required), or major edit category based on the expected editing effort determined using a contour classification model developed in a separate study. The ACC models were trained for the slices requiring minor edit and major edit separately. Performance of the obtained models were tested using an independent 11 MRI sets in term of the change of contour category based on the contour classification model. After applying the duodenum ACC model to the testing datasets, 16% (27/165) and 5% (8/178) of the minor and major edits' slices, respectively, improved to acceptable and 31% (54/178) of the major edit slices improved to minor edits. Furthermore, the total percentage of acceptable contours grew from 10% (36/378) to 19% (71/378) and the percentage of the major edit slices reduced from 47% (178/378) to 30% (115/378). After applying the pancreas ACC model to the testing datasets, 32% (47/143) and 1% (1/96) of the minor and major edits' slices, respectively, improved to acceptable and 49% (47/96) of the major edit slices improved to minor edit slices. Furthermore, the total percentage of acceptable contours grew from 14% (38/277) to 31% (86/277) and the percentage of major edit slices reduced from 35% (96/277) to 17% (48/277). Deep learning based automatic contour corrections can substantially improve inaccurate DLAS contours of pancreas and duodenum on MRI that would otherwise require time-consuming edits, resulting in less manual intervention and increased efficiency during MRgART.

Automatic segmentation of large-scale CT image datasets for detailed body composition analysis

BackgroundBody composition (BC) is an important factor in determining the risk of type 2-diabetes and cardiovascular disease. Computed tomography (CT) is a useful imaging technique for studying BC, however manual segmentation of CT images is time-consuming and subjective. The purpose of this study is to develop and evaluate fully automated segmentation techniques applicable to a 3-slice CT imaging protocol, consisting of single slices at the level of the liver, abdomen, and thigh, allowing detailed analysis of numerous tissues and organs.MethodsThe study used more than 4000 CT subjects acquired from the large-scale SCAPIS and IGT cohort to train and evaluate four convolutional neural network based architectures: ResUNET, UNET++, Ghost-UNET, and the proposed Ghost-UNET++. The segmentation techniques were developed and evaluated for automated segmentation of the liver, spleen, skeletal muscle, bone marrow, cortical bone, and various adipose tissue depots, including visceral (VAT), intraperitoneal (IPAT), retroperitoneal (RPAT), subcutaneous (SAT), deep (DSAT), and superficial SAT (SSAT), as well as intermuscular adipose tissue (IMAT). The models were trained and validated for each target using tenfold cross-validation and test sets.ResultsThe Dice scores on cross validation in SCAPIS were: ResUNET 0.964 (0.909–0.996), UNET++ 0.981 (0.927–0.996), Ghost-UNET 0.961 (0.904–0.991), and Ghost-UNET++ 0.968 (0.910–0.994). All four models showed relatively strong results, however UNET++ had the best performance overall. Ghost-UNET++ performed competitively compared to UNET++ and showed a more computationally efficient approach.ConclusionFully automated segmentation techniques can be successfully applied to a 3-slice CT imaging protocol to analyze multiple tissues and organs related to BC. The overall best performance was achieved by UNET++, against which Ghost-UNET++ showed competitive results based on a more computationally efficient approach. The use of fully automated segmentation methods can reduce analysis time and provide objective results in large-scale studies of BC.

Enhancing teeth segmentation using multifusion deep neural net in panoramic X-ray images.

Precise teeth segmentation from dental panoramic X-ray images is an important task in dental practice. However, several issues including poor image contrast, blurring borders of teeth, presence of jaw bones and other mouth elements, makes reading and examining such images a challenging and time-consuming task for dentists. Thus, developing a precise and automated segmentation technique is required. This study aims to develop and test a novel multi-fusion deep neural net consisting of encoder-decoder architecture for automatic and accurate teeth region segmentation from panoramic X-ray images. The encoder has two different streams based on CNN which include the conventional CNN stream and the Atrous net stream. Next, the fusion of features from these streams is done at each stage to encode the contextual rich information of teeth. A dual-type skip connection is then added between the encoder and decoder to minimise semantic information gaps. Last, the decoder comprises deconvolutional layers for reconstructing the segmented teeth map. The assessment of the proposed model is performed on two different dental datasets consisting of 1,500 and 1,000 panoramic X-ray images, respectively. The new model yields accuracy of 97.0% and 97.7%, intersection over union (IoU) score of 91.1% and 90.2%, and dice coefficient score (DCS) of 92.4% and 90.7% for datasets 1 and 2, respectively. Applying the proposed model to two datasets outperforms the recent state-of-the-art deep models with a relatively smaller number of parameters and higher accuracy, which demonstrates the potential of the new model to help dentists more accurately and efficiently diagnose dental diseases in future clinical practice.

An amalgamation of vision transformer with convolutional neural network for automatic lung tumor segmentation.

Lung cancer has the highest mortality rate. Its diagnosis and treatment analysis depends upon the accurate segmentation of the tumor. It becomes tedious if done manually as radiologists are overburdened with numerous medical imaging tests due to the increase in cancer patients and the COVID pandemic. Automatic segmentation techniques play an essential role in assisting medical experts. The segmentation approaches based on convolutional neural networks have provided state-of-the-art performances. However, they cannot capture long-range relations due to the region-based convolutional operator. Vision Transformers can resolve this issue by capturing global multi-contextual features. To explore this advantageous feature of the vision transformer, we propose an approach for lung tumor segmentation using an amalgamation of the vision transformer and convolutional neural network. We design the network as an encoder-decoder structure with convolution blocks deployed in the initial layers of the encoder to capture the features carrying essential information and the corresponding blocks in the final layers of the decoder. The deeper layers utilize the transformer blocks with a self-attention mechanism to capture more detailed global feature maps. We use a recently proposed unified loss function that combines cross-entropy and dice-based losses for network optimization. We trained our network on a publicly available NSCLC-Radiomics dataset and tested its generalizability on our dataset collected from a local hospital. We could achieve average dice coefficients of 0.7468 and 0.6847 and Hausdorff distances of 15.336 and 17.435 on public and local test data, respectively.

Computer-Aided Diagnosis and Analysis of Skin Cancer from Dermoscopic Images in India.

Researchers have made several advancements in this field, including automatic segmentation techniques, computer-aided diagnosis, mobile-based technology, deep learning methods, hybrid methods etc. All these techniques are beneficial in diagnosing melanoma or segregating skin lesions into different categories. This paper aims to define different types of skin cancers, diagnosis procedures and statistics. This paper presents skin cancer statistics over a period of time in India. The increment in the number of skin carcinoma and melanoma cases from 1990 to 2020 as well as the mortality rates, has been presented in this paper. Also, this paper provides a review of different technologies used by researchers in detecting melanoma. The rise in the number of cases by 2040 and mortality rates are compared. The statistics that are used in this paper are as per hospital-based cancer registries (HBCR) 2021 prepared by the Indian Council of Medical Research - National Centre for Disease Informatics and Research, Bengaluru (ICMR-NCDIR) and from World Health Organization (WHO).

Non-same-scale feature attention network based on BPD for medical image segmentation

The accuracy of diagnosis in medical systems requires automatic image segmentation techniques to provide accurate segmented images of lesions. Segmented images need to be more accurate not only in terms of shape size but also in terms of position. In recent years, a large number of deep learning algorithms have worked tirelessly on this goal. In the field of medical image segmentation, although the prediction images generated by traditional algorithms may not exhibit ideal performance, it is important to note that these methods still provide valuable information regarding edge features. Thus, our goal is to develop a combined approach that integrates traditional algorithms with deep learning techniques. By harnessing the rich edge feature information offered by traditional algorithms, we can enhance the accuracy of image segmentation achieved through deep learning. We propose the Non-same-scale feature attention network based on BPD for medical image segmentation (BPD-NSSFA). First, the network acquires a feature map with rich edge information through Boundary-to-Pixel Direction (BPD) and sends the feature map together with the original image into the backbone network to complete feature extraction and feature fusion. At the bottleneck layer, we use ASPP to expand the receptive field to focus on the associations between more feature information. Finally, we create a Non-same-Scale Feature Attention Block for feature fusion and supervise the fusion process using a deep supervision mechanism. To validate the effectiveness of our network, we select seven different datasets of varying sizes to test the performance of the network. From the experimental results, our network demonstrates superior performance compared to current state-of-the-art methods in lesion localization, edge processing, and noise robustness. Additionally, ablative experiments confirm the rationality of the network structure.

A Deep Learning-Based Automatic Segmentation and 3D Visualization Technique for Intracranial Hemorrhage Detection Using Computed Tomography Images.

Intracranial hemorrhage (ICH) occurs when blood leaks inside the skull as a result of trauma to the skull or due to medical conditions. ICH usually requires immediate medical and surgical attention because the disease has a high mortality rate, long-term disability potential, and other potentially life-threatening complications. There are a wide range of severity levels, sizes, and morphologies of ICHs, making accurate identification challenging. Hemorrhages that are small are more likely to be missed, particularly in healthcare systems that experience high turnover when it comes to computed tomography (CT) investigations. Although many neuroimaging modalities have been developed, CT remains the standard for diagnosing trauma and hemorrhage (including non-traumatic ones). A CT scan-based diagnosis can provide time-critical, urgent ICH surgery that could save lives because CT scan-based diagnoses can be obtained rapidly. The purpose of this study is to develop a machine-learning algorithm that can detect intracranial hemorrhage based on plain CT images taken from 75 patients. CT images were preprocessed using brain windowing, skull-stripping, and image inversion techniques. Hemorrhage segmentation was performed using multiple pre-trained models on preprocessed CT images. A U-Net model with DenseNet201 pre-trained encoder outperformed other U-Net, U-Net++, and FPN (Feature Pyramid Network) models with the highest Dice similarity coefficient (DSC) and intersection over union (IoU) scores, which were previously used in many other medical applications. We presented a three-dimensional brain model highlighting hemorrhages from ground truth and predicted masks. The volume of hemorrhage was measured volumetrically to determine the size of the hematoma. This study is essential in examining ICH for diagnostic purposes in clinical practice by comparing the predicted 3D model with the ground truth.

Using a modified U-net model to improve Retinal Vessel Segmentation

AI applications in healthcare have broad and exciting prospects, especially in medical image segmentation where AI has proved it can do a better job than human beings in specific tasks. In this project, we apply a U-net model to complete the retinal vessel segmentation task and compare the results with the morphology method which is a classical automatic segmentation technique. Besides, we also make an attempt at data enhancement in morphology methods to explore the importance of data preprocessing in digital image processing. The result shows that the U-net model produces better outcomes than the morphology method according to most of the evaluation criteria. However, data enhancement in morphology does not show significant differences from the original.