Early detection and stratification of colorectal cancer using plasma cell-free DNA fragmentomic profiling.

Timely accurate and cost-efficient detection of colorectal cancer (CRC) is of great clinical importance. This study aims to establish prediction models for detecting CRC using plasma cell-free DNA (cfDNA) fragmentomic features. Whole-genome sequencing (WGS) was performed on cfDNA from 620 participants, including healthy individuals, patients with benign colorectal diseases and CRC patients. Using WGS data, three machine learning methods were compared to build prediction models for the stratification of CRC patients. The optimal model to discriminate CRC patients of all stages from healthy individuals achieved a sensitivity of 92.31% and a specificity of 91.14%, while the model to separate early-stage CRC patients (stage 0-II) from healthy individuals achieved a sensitivity of 88.8% and a specificity of 96.2%. Additionally, the cfDNA fragmentation profiles reflected disease-specific genomic alterations in CRC. Overall, this study suggests that cfDNA fragmentation profiles may potentially become a noninvasive approach for the detection and stratification of CRC.

NanoBERTa-ASP: predicting nanobody paratope based on a pretrained RoBERTa model.

BACKGROUND: Nanobodies, also known as VHH or single-domain antibodies, are unique antibody fragments derived solely from heavy chains. They offer advantages of small molecules and conventional antibodies, making them promising therapeutics. The paratope is the specific region on an antibody that binds to an antigen. Paratope prediction involves the identification and characterization of the antigen-binding site on an antibody. This process is crucial for understanding the specificity and affinity of antibody-antigen interactions. Various computational methods and experimental approaches have been developed to predict and analyze paratopes, contributing to advancements in antibody engineering, drug development, and immunotherapy. However, existing predictive models trained on traditional antibodies may not be suitable for nanobodies. Additionally, the limited availability of nanobody datasets poses challenges in constructing accurate models. METHODS: To address these challenges, we have developed a novel nanobody prediction model, named NanoBERTa-ASP (Antibody Specificity Prediction), which is specifically designed for predicting nanobody-antigen binding sites. The model adopts a training strategy more suitable for nanobodies, based on an advanced natural language processing (NLP) model called BERT (Bidirectional Encoder Representations from Transformers). To be more specific, the model utilizes a masked language modeling approach named RoBERTa (Robustly Optimized BERT Pretraining Approach) to learn the contextual information of the nanobody sequence and predict its binding site. RESULTS: NanoBERTa-ASP achieved exceptional performance in predicting nanobody binding sites, outperforming existing methods, indicating its proficiency in capturing sequence information specific to nanobodies and accurately identifying their binding sites. Furthermore, NanoBERTa-ASP provides insights into the interaction mechanisms between nanobodies and antigens, contributing to a better understanding of nanobodies and facilitating the design and development of nanobodies with therapeutic potential. CONCLUSION: NanoBERTa-ASP represents a significant advancement in nanobody paratope prediction. Its superior performance highlights the potential of deep learning approaches in nanobody research. By leveraging the increasing volume of nanobody data, NanoBERTa-ASP can further refine its predictions, enhance its performance, and contribute to the development of novel nanobody-based therapeutics. Github repository: https://github.com/WangLabforComputationalBiology/NanoBERTa-ASP.

Detection and monitoring of HBV-related hepatocellular carcinoma from plasma cfDNA fragmentation profiles.

Most hepatocellular carcinomas (HCCs) are associated with hepatitis B virus infection (HBV) in China. Early detection of HCC can significantly improve prognosis but is not yet fully clinically feasible. This study aims to develop methods for detecting HCC and studying the carcinogenesis of HBV using plasma cell-free DNA (cfDNA) whole-genome sequencing (WGS) data. Low coverage WGS was performed for 452 participants, including healthy individuals, hepatitis B patients, cirrhosis patients, and HCC patients. Then the sequencing data were processed using various machine learning models based on cfDNA fragmentation profiles for cancer detection. Our best model achieved a sensitivity of 87.10% and a specificity of 88.37%, and it showed an increased sensitivity with higher BCLC stages of HCC. Overall, this study proves the potential of a non-invasive assay based on cfDNA fragmentation profiles for the detection and prognosis of HCC and provides preliminary data on the carcinogenic mechanism of HBV.

BreakID: genomics breakpoints identification to detect gene fusion events using discordant pairs and split reads.

SUMMARY: Here we developed a tool called Breakpoint Identification (BreakID) to identity fusion events from targeted sequencing data. Taking discordant read pairs and split reads as supporting evidences, BreakID can identify gene fusion breakpoints at single nucleotide resolution. After validation with confirmed fusion events in cancer cell lines, we have proved that BreakID can achieve high sensitivity of 90.63% along with PPV of 100% at sequencing depth of 500× and perform better than other available fusion detection tools. We anticipate that BreakID will have an extensive popularity in the detection and analysis of fusions involved in clinical and research sequencing scenarios. AVAILABILITY AND IMPLEMENTATION: Source code is freely available at https://github.com/SinOncology/BreakID. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The conserved Lys-95 charged residue cluster is critical for the homodimerization and enzyme activity of human ribonucleotide reductase small subunit M2.

Ribonucleotide reductase (RR) catalyzes the reduction of ribonucleotides to deoxyribonucleotides for DNA synthesis. Human RR small subunit M2 exists in a homodimer form. However, the importance of the dimer form to the enzyme and the related mechanism remain unclear. In this study, we tried to identify the interfacial residues that may mediate the assembly of M2 homodimer by computational alanine scanning based on the x-ray crystal structure. Co-immunoprecipitation, size exclusion chromatography, and RR activity assays showed that the K95E mutation in M2 resulted in dimer disassembly and enzyme activity inhibition. In comparison, the charge-exchanging double mutation of K95E and E98K recovered the dimerization and activity. Structural comparisons suggested that a conserved cluster of charged residues, including Lys-95, Glu-98, Glu-105, and Glu-174, at the interface may function as an ionic lock for M2 homodimer. Although the measurements of the radical and iron contents showed that the monomer (the K95E mutant) was capable of generating the diiron and tyrosyl radical cofactor, co-immunoprecipitation and competitive enzyme inhibition assays indicated that the disassembly of M2 dimer reduced its interaction with the large subunit M1. In addition, the immunofluorescent and fusion protein-fluorescent imaging analyses showed that the dissociation of M2 dimer altered its subcellular localization. Finally, the transfection of the wild-type M2 but not the K95E mutant rescued the G1/S phase cell cycle arrest and cell growth inhibition caused by the siRNA knockdown of M2. Thus, the conserved Lys-95 charged residue cluster is critical for human RR M2 homodimerization, which is indispensable to constitute an active holoenzyme and function in cells.

NanoLAS: a comprehensive nanobody database with data integration, consolidation and application.

Nanobodies, a unique subclass of antibodies first discovered in camelid animals, are composed solely of a single heavy chain's variable region. Their significantly reduced molecular weight, in comparison to conventional antibodies, confers numerous advantages in the treatment of various diseases. As research and applications involving nanobodies expand, the quantity of identified nanobodies is also rapidly growing. However, the existing antibody databases are deficient in type and coverage, failing to satisfy the comprehensive needs of researchers and thus impeding progress in nanobody research. In response to this, we have amalgamated data from multiple sources to successfully assemble a new and comprehensive nanobody database. This database has currently included the latest nanobody data and provides researchers with an excellent search and data display interface, thus facilitating the progression of nanobody research and their application in disease treatment. In summary, the newly constructed Nanobody Library and Archive System may significantly enhance the retrieval efficiency and application potential of nanobodies. We envision that Nanobody Library and Archive System will serve as an accessible, robust and efficient tool for nanobody research and development, propelling advancements in the field of biomedicine. Database URL: https://www.nanolas.cloud.

Customized m-RCNN and hybrid deep classifier for liver cancer segmentation and classification.

Diagnosing liver disease presents a significant medical challenge in impoverished countries, with over 30 billion individuals succumbing to it each year. Existing models for detecting liver abnormalities suffer from lower accuracy and higher constraint metrics. As a result, there is a pressing need for improved, efficient, and effective liver disease detection methods. To address the limitations of current models, this method introduces a deep liver segmentation and classification system based on a Customized Mask-Region Convolutional Neural Network (cm-RCNN). The process begins with preprocessing the input liver image, which includes Adaptive Histogram Equalization (AHE). AHE helps dehaze the input image, remove color distortion, and apply linear transformations to obtain the preprocessed image. Next, a precise region of interest is segmented from the preprocessed image using a novel deep strategy called cm-RCNN. To enhance segmentation accuracy, the architecture incorporates the ReLU activation function and the modified sigmoid activation function. Subsequently, a variety of features are extracted from the segmented image, including ResNet features, shape features (area, perimeter, approximation, and convex hull), and enhanced median binary pattern. These extracted features are then used to train a hybrid classification model, which incorporates classifiers like SqueezeNet and DeepMaxout models. The final classification outcome is determined by averaging the scores obtained from both classifiers.

Iterative In Silico Screening for Optimizing Stable Conformation of Anti-SARS-CoV-2 Nanobodies.

Nanobodies (Nbs or VHHs) are single-domain antibodies (sdAbs) derived from camelid heavy-chain antibodies. Nbs have special and unique characteristics, such as small size, good tissue penetration, and cost-effective production, making Nbs a good candidate for the diagnosis and treatment of viruses and other pathologies. Identifying effective Nbs against COVID-19 would help us control this dangerous virus or other unknown variants in the future. Herein, we introduce an in silico screening strategy for optimizing stable conformation of anti-SARS-CoV-2 Nbs. Firstly, various complexes containing nanobodies were downloaded from the RCSB database, which were identified from immunized llamas. The primary docking between Nbs and the SARS-CoV-2 spike protein receptor-binding domain was performed through the ClusPro program, with the manual screening leaving the reasonable conformation to the next step. Then, the binding distances of atoms between the antigen-antibody interfaces were measured through the NeighborSearch algorithm. Finally, filtered nanobodies were acquired according to HADDOCK scores through HADDOCK docking the COVID-19 spike protein with nanobodies under restrictions of calculated molecular distance between active residues and antigenic epitopes less than 4.5 Å. In this way, those nanobodies with more reasonable conformation and stronger neutralizing efficacy were acquired. To validate the efficacy ranking of the nanobodies we obtained, we calculated the binding affinities (∆G) and dissociation constants (Kd) of all screened nanobodies using the PRODIGY web tool and predicted the stability changes induced by all possible point mutations in nanobodies using the MAESTROWeb server. Furthermore, we examined the performance of the relationship between nanobodies' ranking and their number of mutation-sensitive sites (Spearman correlation > 0.68); the results revealed a robust correlation, indicating that the superior nanobodies identified through our screening process exhibited fewer mutation hotspots and higher stability. This correlation analysis demonstrates the validity of our screening criteria, underscoring the suitability of these nanobodies for future development and practical implementation. In conclusion, this three-step screening strategy iteratively in silico greatly improved the accuracy of screening desired nanobodies compared to using only ClusPro docking or default HADDOCK docking settings. It provides new ideas for the screening of novel antibodies and computer-aided screening methods.

Transformative Deep Neural Network Approaches in Kidney Ultrasound Segmentation: Empirical Validation with an Annotated Dataset.

Kidney ultrasound (US) images are primarily employed for diagnosing different renal diseases. Among them, one is renal localization and detection, which can be carried out by segmenting the kidney US images. However, kidney segmentation from US images is challenging due to low contrast, speckle noise, fluid, variations in kidney shape, and modality artifacts. Moreover, well-annotated US datasets for renal segmentation and detection are scarce. This study aims to build a novel, well-annotated dataset containing 44,880 US images. In addition, we propose a novel training scheme that utilizes the encoder and decoder parts of a state-of-the-art segmentation algorithm. In the pre-processing step, pixel intensity normalization improves contrast and facilitates model convergence. The modified encoder-decoder architecture improves pyramid-shaped hole pooling, cascaded multiple-hole convolutions, and batch normalization. The pre-processing step gradually reconstructs spatial information, including the capture of complete object boundaries, and the post-processing module with a concave curvature reduces the false positive rate of the results. We present benchmark findings to validate the quality of the proposed training scheme and dataset. We applied six evaluation metrics and several baseline segmentation approaches to our novel kidney US dataset. Among the evaluated models, DeepLabv3+ performed well and achieved the highest dice, Hausdorff distance 95, accuracy, specificity, average symmetric surface distance, and recall scores of 89.76%, 9.91, 98.14%, 98.83%, 3.03, and 90.68%, respectively. The proposed training strategy aids state-of-the-art segmentation models, resulting in better-segmented predictions. Furthermore, the large, well-annotated kidney US public dataset will serve as a valuable baseline source for future medical image analysis research.

Adaptive Feature Medical Segmentation Network: an adaptable deep learning paradigm for high-performance 3D brain lesion segmentation in medical imaging.

Introduction: In neurological diagnostics, accurate detection and segmentation of brain lesions is crucial. Identifying these lesions is challenging due to its complex morphology, especially when using traditional methods. Conventional methods are either computationally demanding with a marginal impact/enhancement or sacrifice fine details for computational efficiency. Therefore, balancing performance and precision in compute-intensive medical imaging remains a hot research topic. Methods: We introduce a novel encoder-decoder network architecture named the Adaptive Feature Medical Segmentation Network (AFMS-Net) with two encoder variants: the Single Adaptive Encoder Block (SAEB) and the Dual Adaptive Encoder Block (DAEB). A squeeze-and-excite mechanism is employed in SAEB to identify significant data while disregarding peripheral details. This approach is best suited for scenarios requiring quick and efficient segmentation, with an emphasis on identifying key lesion areas. In contrast, the DAEB utilizes an advanced channel spatial attention strategy for fine-grained delineation and multiple-class classifications. Additionally, both architectures incorporate a Segmentation Path (SegPath) module between the encoder and decoder, refining segmentation, enhancing feature extraction, and improving model performance and stability. Results: AFMS-Net demonstrates exceptional performance across several notable datasets, including BRATs 2021, ATLAS 2021, and ISLES 2022. Its design aims to construct a lightweight architecture capable of handling complex segmentation challenges with high precision. Discussion: The proposed AFMS-Net addresses the critical balance issue between performance and computational efficiency in the segmentation of brain lesions. By introducing two tailored encoder variants, the network adapts to varying requirements of speed and feature. This approach not only advances the state-of-the-art in lesion segmentation but also provides a scalable framework for future research in medical image processing.

Gut microbial co-metabolite 2-methylbutyrylcarnitine exacerbates thrombosis via binding to and activating integrin α2ß1.

Thrombosis represents the leading cause of death and disability upon major adverse cardiovascular events (MACEs). Numerous pathological conditions such as COVID-19 and metabolic disorders can lead to a heightened thrombotic risk; however, the underlying mechanisms remain poorly understood. Our study illustrates that 2-methylbutyrylcarnitine (2MBC), a branched-chain acylcarnitine, is accumulated in patients with COVID-19 and in patients with MACEs. 2MBC enhances platelet hyperreactivity and thrombus formation in mice. Mechanistically, 2MBC binds to integrin α2ß1 in platelets, potentiating cytosolic phospholipase A2 (cPLA2) activation and platelet hyperresponsiveness. Genetic depletion or pharmacological inhibition of integrin α2ß1 largely reverses the pro-thrombotic effects of 2MBC. Notably, 2MBC can be generated in a gut-microbiota-dependent manner, whereas the accumulation of plasma 2MBC and its thrombosis-aggravating effect are largely ameliorated following antibiotic-induced microbial depletion. Our study implicates 2MBC as a metabolite that links gut microbiota dysbiosis to elevated thrombotic risk, providing mechanistic insight and a potential therapeutic strategy for thrombosis.

Representation Learning and Reinforcement Learning for Dynamic Complex Motion Planning System.

Indoor motion planning challenges researchers because of the high density and unpredictability of moving obstacles. Classical algorithms work well in the case of static obstacles but suffer from collisions in the case of dense and dynamic obstacles. Recent reinforcement learning (RL) algorithms provide safe solutions for multiagent robotic motion planning systems. However, these algorithms face challenges in convergence: slow convergence speed and suboptimal converged result. Inspired by RL and representation learning, we introduced the ALN-DSAC: a hybrid motion planning algorithm where attention-based long short-term memory (LSTM) and novel data replay combine with discrete soft actor-critic (SAC). First, we implemented a discrete SAC algorithm, which is the SAC in the setting of discrete action space. Second, we optimized existing distance-based LSTM encoding by attention-based encoding to improve the data quality. Third, we introduced a novel data replay method by combining the online learning and offline learning to improve the efficacy of data replay. The convergence of our ALN-DSAC outperforms that of the trainable state of the arts. Evaluations demonstrate that our algorithm achieves nearly 100% success with less time to reach the goal in motion planning tasks when compared to the state of the arts. The test code is available at https://github.com/CHUENGMINCHOU/ALN-DSAC.

A Robust Brain Tumor Detector Using BiLSTM and Mayfly Optimization and Multi-Level Thresholding.

A brain tumor refers to an abnormal growth of cells in the brain that can be either benign or malignant. Oncologists typically use various methods such as blood or visual tests to detect brain tumors, but these approaches can be time-consuming, require additional human effort, and may not be effective in detecting small tumors. This work proposes an effective approach to brain tumor detection that combines segmentation and feature fusion. Segmentation is performed using the mayfly optimization algorithm with multilevel Kapur's threshold technique to locate brain tumors in MRI scans. Key features are achieved from tumors employing Histogram of Oriented Gradients (HOG) and ResNet-V2, and a bidirectional long short-term memory (BiLSTM) network is used to classify tumors into three categories: pituitary, glioma, and meningioma. The suggested methodology is trained and tested on two datasets, Figshare and Harvard, achieving high accuracy, precision, recall, F1 score, and area under the curve (AUC). The results of a comparative analysis with existing DL and ML methods demonstrate that the proposed approach offers superior outcomes. This approach has the potential to improve brain tumor detection, particularly for small tumors, but further validation and testing are needed before clinical use.

Optical navigation robot-assisted puncture system for accurate lung nodule biopsy: an animal study.

Background: As lung cancer is one of the most significant factors seriously endangering human health, a robot-assisted puncture system with high accuracy and safety is urgently needed. The purpose of this investigation was to compare the safety and effectiveness of such a robot-assisted system to the conventional computed tomography (CT)-guided manual method for percutaneous lung biopsies (PLBs) in pigs. Methods: An optical navigation robot-assisted puncture system was developed and compared to the traditional CT-guided PLB using simulated lesions in experimental animals. A total of 30 pulmonary nodules were successfully created in 5 pigs (Wuzhishan pig, 1 male and 4 females). Of these, 15 were punctured by the optical navigation robot-assisted puncture system (robotic group), and 15 were manually punctured under CT guidance (manual group). The biopsy success rate, operation time, first needle tip-target point deviation, and needle adjustment times were compared between groups. Postoperative CT scans were performed to identify complications. Results: The single puncture success rate was higher in the robotic group (13/15; 86.7%) than in the manual group (8/15; 53.3%). The first puncture was closer to the target lesion (1.8±1.7 mm), and the operation time was shorter (7.1±3.7 minutes) in the robotic group than in the manual group (4.4±2.8 mm and 12.9±7.6 minutes, respectively). The angle deviation was smaller in the robotic group (3.26°±2.48°) than in the manual group (7.71°±3.86°). The robotic group displayed significant advantages (P<0.05). The primary complication in both groups was slight bleeding, with an incidence of 26.7% in the robotic group and 40.0% in the manual group. There was 1 case of pneumothorax in the manual group, and there were no deaths due to complications in either group. Conclusions: An optical navigation robot-assisted system for PLBs guided by CT images was developed and demonstrated. The experimental results indicate that the proposed system is accurate, efficient, and safe in pigs.

Identification of cavities on protein surface using multiple computational approaches for drug binding site prediction.

MOTIVATION: Protein-ligand binding sites are the active sites on protein surface that perform protein functions. Thus, the identification of those binding sites is often the first step to study protein functions and structure-based drug design. There are many computational algorithms and tools developed in recent decades, such as LIGSITE(cs/c), PASS, Q-SiteFinder, SURFNET, and so on. In our previous work, MetaPocket, we have proved that it is possible to combine the results of many methods together to improve the prediction result. RESULTS: Here, we continue our previous work by adding four more methods Fpocket, GHECOM, ConCavity and POCASA to further improve the prediction success rate. The new method MetaPocket 2.0 and the individual approaches are all tested on two datasets of 48 unbound/bound and 210 bound structures as used before. The results show that the average success rate has been raised 5% at the top 1 prediction compared with previous work. Moreover, we construct a non-redundant dataset of drug-target complexes with known structure from DrugBank, DrugPort and PDB database and apply MetaPocket 2.0 to this dataset to predict drug binding sites. As a result, >74% drug binding sites on protein target are correctly identified at the top 3 prediction, and it is 12% better than the best individual approach. AVAILABILITY: The web service of MetaPocket 2.0 and all the test datasets are freely available at http://projects.biotec.tu-dresden.de/metapocket/ and http://sysbio.zju.edu.cn/metapocket.

Deep Learning Applications in Computed Tomography Images for Pulmonary Nodule Detection and Diagnosis: A Review.

Lung cancer has one of the highest mortality rates of all cancers and poses a severe threat to people's health. Therefore, diagnosing lung nodules at an early stage is crucial to improving patient survival rates. Numerous computer-aided diagnosis (CAD) systems have been developed to detect and classify such nodules in their early stages. Currently, CAD systems for pulmonary nodules comprise data acquisition, pre-processing, lung segmentation, nodule detection, false-positive reduction, segmentation, and classification. A number of review articles have considered various components of such systems, but this review focuses on segmentation and classification parts. Specifically, categorizing segmentation parts based on lung nodule type and network architectures, i.e., general neural network and multiview convolution neural network (CNN) architecture. Moreover, this work organizes related literature for classification of parts based on nodule or non-nodule and benign or malignant. The essential CT lung datasets and evaluation metrics used in the detection and diagnosis of lung nodules have been systematically summarized as well. Thus, this review provides a baseline understanding of the topic for interested readers.

Supervised and weakly supervised deep learning models for COVID-19 CT diagnosis: A systematic review.

Artificial intelligence (AI) and computer vision (CV) methods become reliable to extract features from radiological images, aiding COVID-19 diagnosis ahead of the pathogenic tests and saving critical time for disease management and control. Thus, this review article focuses on cascading numerous deep learning-based COVID-19 computerized tomography (CT) imaging diagnosis research, providing a baseline for future research. Compared to previous review articles on the topic, this study pigeon-holes the collected literature very differently (i.e., its multi-level arrangement). For this purpose, 71 relevant studies were found using a variety of trustworthy databases and search engines, including Google Scholar, IEEE Xplore, Web of Science, PubMed, Science Direct, and Scopus. We classify the selected literature in multi-level machine learning groups, such as supervised and weakly supervised learning. Our review article reveals that weak supervision has been adopted extensively for COVID-19 CT diagnosis compared to supervised learning. Weakly supervised (conventional transfer learning) techniques can be utilized effectively for real-time clinical practices by reusing the sophisticated features rather than over-parameterizing the standard models. Few-shot and self-supervised learning are the recent trends to address data scarcity and model efficacy. The deep learning (artificial intelligence) based models are mainly utilized for disease management and control. Therefore, it is more appropriate for readers to comprehend the related perceptive of deep learning approaches for the in-progress COVID-19 CT diagnosis research.

Towards more efficient ophthalmic disease classification and lesion location via convolution transformer.

OBJECTIVE: A retina optical coherence tomography (OCT) image differs from a traditional image due to its significant speckle noise, irregularity, and inconspicuous features. A conventional deep learning architecture cannot effectively improve the classification accuracy, sensitivity, and specificity of OCT images, and noisy images are not conducive to further diagnosis.  This paper proposes a novel lesion-localization convolution transformer (LLCT) method, which combines both convolution and self-attention to classify ophthalmic diseases more accurately and localize the lesions in retina OCT images. METHODS: A novel architecture design is accomplished through applying customized feature maps generated by convolutional neutral network (CNN) as the input sequence of self-attention network. This design takes advantages of CNN's extracting image features and transformer's consideration of global context and dynamic attention. Part of the model is backward propagated to calculate the gradient as a weight parameter, which is multiplied and summed with the global features generated by the forward propagation process to locate the lesion. RESULTS: Extensive experiments show that our proposed design achieves improvement of about 7.6% in overall accuracy, 10.9% in overall sensitivity, and 9.2% in overall specificity compared with previous methods. And the lesions can be localized without the labeling data of lesion location in OCT images. CONCLUSION: The results prove that our method significantly improves the performance and reduces the computation complexity in artificial intelligence assisted analysis of ophthalmic disease through OCT images. SIGNIFICANCE: Our method has a significance boost in ophthalmic disease classification and location via convolution transformer. This is applicable to assist ophthalmologists greatly.1.

Deep Segmentation Networks for Segmenting Kidneys and Detecting Kidney Stones in Unenhanced Abdominal CT Images.

Recent breakthroughs of deep learning algorithms in medical imaging, automated detection, and segmentation techniques for renal (kidney) in abdominal computed tomography (CT) images have been limited. Radiomics and machine learning analyses of renal diseases rely on the automatic segmentation of kidneys in CT images. Inspired by this, our primary aim is to utilize deep semantic segmentation learning models with a proposed training scheme to achieve precise and accurate segmentation outcomes. Moreover, this work aims to provide the community with an open-source, unenhanced abdominal CT dataset for training and testing the deep learning segmentation networks to segment kidneys and detect kidney stones. Five variations of deep segmentation networks are trained and tested both dependently (based on the proposed training scheme) and independently. Upon comparison, the models trained with the proposed training scheme enable the highly accurate 2D and 3D segmentation of kidneys and kidney stones. We believe this work is a fundamental step toward AI-driven diagnostic strategies, which can be an essential component of personalized patient care and improved decision-making in treating kidney diseases.