Mulberry leaf dataset for image classification task.

This manuscript presents a mulberry leaf dataset collected from five provinces within three regions in Thailand. The dataset contains ten categories of mulberry leaves. We proposed this dataset due to the challenges of classifying leaf images taken in natural environments arising from high inter-class similarity and variations in illumination and background conditions (multiple leaves from a mulberry tree and shadows appearing in the leaf images). We highlight that our research team recorded mulberry leaves independently from various perspectives during our data acquisition using multiple camera types. The mulberry leaf dataset can serve as vital input data passed to computer vision algorithms (conventional deep learning and vision transformer algorithms) for creating image recognition systems. The dataset will allow other researchers to propose novel computer vision techniques to approach mulberry recognition challenges.

Integration between constrained optimization and deep networks: a survey.

Integration between constrained optimization and deep networks has garnered significant interest from both research and industrial laboratories. Optimization techniques can be employed to optimize the choice of network structure based not only on loss and accuracy but also on physical constraints. Additionally, constraints can be imposed during training to enhance the performance of networks in specific contexts. This study surveys the literature on the integration of constrained optimization with deep networks. Specifically, we examine the integration of hyper-parameter tuning with physical constraints, such as the number of FLOPS (FLoating point Operations Per Second), a measure of computational capacity, latency, and other factors. This study also considers the use of context-specific knowledge constraints to improve network performance. We discuss the integration of constraints in neural architecture search (NAS), considering the problem as both a multi-objective optimization (MOO) challenge and through the imposition of penalties in the loss function. Furthermore, we explore various approaches that integrate logic with deep neural networks (DNNs). In particular, we examine logic-neural integration through constrained optimization applied during the training of NNs and the use of semantic loss, which employs the probabilistic output of the networks to enforce constraints on the output.

Revolutionizing Healthcare: Qure.AI's Innovations in Medical Diagnosis and Treatment.

Qure.AI, a leading company in artificial intelligence (AI) applied to healthcare, has developed a suite of innovative solutions to revolutionize medical diagnosis and treatment. With a plethora of FDA-approved tools for clinical use, Qure.AI continually strives for innovation in integrating AI into healthcare systems. This article delves into the efficacy of Qure.AI's chest X-ray interpretation tool, "qXR," in medicine, drawing from a comprehensive review of clinical trials conducted by various institutions. Key applications of AI in healthcare include machine learning, deep learning, and natural language processing (NLP), all of which contribute to enhanced diagnostic accuracy, efficiency, and speed. Through the analysis of vast datasets, AI algorithms assist physicians in interpreting medical data and making informed decisions, thereby improving patient care outcomes. Illustrative examples highlight AI's impact on medical imaging, particularly in the diagnosis of conditions such as breast cancer, heart failure, and pulmonary nodules. AI can significantly reduce diagnostic errors and expedite the interpretation of medical images, leading to more timely interventions and treatments. Furthermore, AI-powered predictive analytics enable early detection of diseases and facilitate personalized treatment plans, thereby reducing healthcare costs and improving patient outcomes. The efficacy of AI in healthcare is underscored by its ability to complement traditional diagnostic methods, providing physicians with valuable insights and support in clinical decision-making. As AI continues to evolve, its role in patient care and medical research is poised to expand, promising further advancements in diagnostic accuracy and treatment efficacy.

Comparison of image quality between Deep learning image reconstruction and Iterative reconstruction technique for CT Brain- a pilot study.

Background: Non-contrast Computed Tomography (NCCT) plays a pivotal role in assessing central nervous system disorders and is a crucial diagnostic method. Iterative reconstruction (IR) methods have enhanced image quality (IQ) but may result in a blotchy appearance and decreased resolution for subtle contrasts. The deep-learning image reconstruction (DLIR) algorithm, which integrates a convolutional neural network (CNN) into the reconstruction process, generates high-quality images with minimal noise. Hence, the objective of this study was to assess the IQ of the Precise Image (DLIR) and the IR technique (iDose 4) for the NCCT brain. Methods: This is a prospective study. Thirty patients who underwent NCCT brain were included. The images were reconstructed using DLIR-standard and iDose 4. Qualitative IQ analysis parameters, such as overall image quality (OQ), subjective image noise (SIN), and artifacts, were measured. Quantitative IQ analysis parameters such as Computed Tomography (CT) attenuation (HU), image noise (IN), posterior fossa index (PFI), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) in the basal ganglia (BG) and centrum-semiovale (CSO) were measured. Paired t-tests were performed for qualitative and quantitative IQ analyses between the iDose 4 and DLIR-standard. Kappa statistics were used to assess inter-observer agreement for qualitative analysis. Results: Quantitative IQ analysis showed significant differences (p<0.05) in IN, SNR, and CNR between the iDose 4 and DLIR-standard at the BG and CSO levels. IN was reduced (41.8-47.6%), SNR (65-82%), and CNR (68-78.8%) were increased with DLIR-standard. PFI was reduced (27.08%) the DLIR-standard. Qualitative IQ analysis showed significant differences (p<0.05) in OQ, SIN, and artifacts between the DLIR standard and iDose 4. The DLIR standard showed higher qualitative IQ scores than the iDose 4. Conclusion: DLIR standard yielded superior quantitative and qualitative IQ compared to the IR technique (iDose4). The DLIR-standard significantly reduced the IN and artifacts compared to iDose 4 in the NCCT brain.

Dual-stream multi-dependency graph neural network enables precise cancer survival analysis.

Histopathology image-based survival prediction aims to provide a precise assessment of cancer prognosis and can inform personalized treatment decision-making in order to improve patient outcomes. However, existing methods cannot automatically model the complex correlations between numerous morphologically diverse patches in each whole slide image (WSI), thereby preventing them from achieving a more profound understanding and inference of the patient status. To address this, here we propose a novel deep learning framework, termed dual-stream multi-dependency graph neural network (DM-GNN), to enable precise cancer patient survival analysis. Specifically, DM-GNN is structured with the feature updating and global analysis branches to better model each WSI as two graphs based on morphological affinity and global co-activating dependencies. As these two dependencies depict each WSI from distinct but complementary perspectives, the two designed branches of DM-GNN can jointly achieve the multi-view modeling of complex correlations between the patches. Moreover, DM-GNN is also capable of boosting the utilization of dependency information during graph construction by introducing the affinity-guided attention recalibration module as the readout function. This novel module offers increased robustness against feature perturbation, thereby ensuring more reliable and stable predictions. Extensive benchmarking experiments on five TCGA datasets demonstrate that DM-GNN outperforms other state-of-the-art methods and offers interpretable prediction insights based on the morphological depiction of high-attention patches. Overall, DM-GNN represents a powerful and auxiliary tool for personalized cancer prognosis from histopathology images and has great potential to assist clinicians in making personalized treatment decisions and improving patient outcomes.

Deep learning in sex estimation from photographed human mandible using the Human Osteological Research Collection.

Sex estimation is a necessary part of forensic and osteological analyses of skeletal human remains in the construction of a biological profile. Several skeletal traits are sexually dimorphic and used for skeletal sex estimation. The human mandible and morphological traits therein have been long used for sex estimation, but the validity of using the mandible in this purpose has become a concern. In this study, we examined the potential of artificial intelligence (AI) and especially deep learning (DL) to provide accurate sex estimations from the mandible. We used 193 modern South African mandibles from the Human Osteological Research Collection (HORC) in the Sefako Makgatho Health Sciences university with known sex to conduct our study. All mandibles were photographed from the same angle and the photographs were analyzed with an open-source DL software. The best-performing DL algorithm estimated the sex of males with 100% accuracy and females with 76.9% accuracy. However, further studies with a higher number of specimens could provide more reliable validity for using AI when building the biological profile from skeletal remains.

Prospective multicenter study of continuous tonic-clonic seizure monitoring on Apple Watch in epilepsy monitoring units and ambulatory environments.

OBJECTIVE: Evaluate the performance of a custom application developed for tonic-clonic seizure (TCS) monitoring on a consumer-wearable (Apple Watch) device. METHODS: Participants with a history of convulsive epileptic seizures were recruited for either Epilepsy Monitoring Unit (EMU) or ambulatory (AMB) monitoring; participants without epilepsy (normal controls [NC]) were also enrolled in the AMB group. Both EMU and AMB participants wore an Apple Watch with a research app that continuously recorded accelerometer and photoplethysmography (PPG) signals, and ran a fixed-and-frozen tonic-clonic seizure detection algorithm during the testing period. This algorithm had been previously developed and validated using a separate training dataset. All EMU convulsive events were validated by video-electroencephalography (video-EEG); AMB events were validated by caregiver reporting and follow-ups. Device performance was characterized and compared to prior monitoring devices through sensitivity, false alarm rate (FAR; false-alarms per 24 h), precision, and detection delay (latency). RESULTS: The EMU group had 85 participants (4,279 h, 19 TCS from 15 participants) enrolled across four EMUs; the AMB group had 21 participants (13 outpatient, 8 NC, 6,735 h, 10 TCS from 3 participants). All but one AMB participant completed the study. Device performance in the EMU group included a sensitivity of 100 % [95 % confidence interval (CI) 79-100 %]; an FAR of 0.05 [0.02, 0.08] per 24 h; a precision of 68 % [48 %, 83 %]; and a latency of 32.07 s [standard deviation (std) 10.22 s]. The AMB group had a sensitivity of 100 % [66-100 %]; an FAR of 0.13 [0.08, 0.24] per 24 h; a precision of 22 % [11 %, 37 %]; and a latency of 37.38 s [13.24 s]. Notably, a single AMB participant was responsible for 8 of 31 false alarms. The AMB FAR excluding this participant was 0.10 [0.07, 0.14] per 24 h. DISCUSSION: This study demonstrates the practicability of TCS monitoring on a popular consumer wearable (Apple Watch) in daily use for people with epilepsy. The monitoring app had a high sensitivity and a substantially lower FAR than previously reported in both EMU and AMB environments.

Current strategies to address data scarcity in artificial intelligence-based drug discovery: A comprehensive review.

Artificial intelligence (AI) has played a vital role in computer-aided drug design (CADD). This development has been further accelerated with the increasing use of machine learning (ML), mainly deep learning (DL), and computing hardware and software advancements. As a result, initial doubts about the application of AI in drug discovery have been dispelled, leading to significant benefits in medicinal chemistry. At the same time, it is crucial to recognize that AI is still in its infancy and faces a few limitations that need to be addressed to harness its full potential in drug discovery. Some notable limitations are insufficient, unlabeled, and non-uniform data, the resemblance of some AI-generated molecules with existing molecules, unavailability of inadequate benchmarks, intellectual property rights (IPRs) related hurdles in data sharing, poor understanding of biology, focus on proxy data and ligands, lack of holistic methods to represent input (molecular structures) to prevent pre-processing of input molecules (feature engineering), etc. The major component in AI infrastructure is input data, as most of the successes of AI-driven efforts to improve drug discovery depend on the quality and quantity of data, used to train and test AI algorithms, besides a few other factors. Additionally, data-gulping DL approaches, without sufficient data, may collapse to live up to their promise. Current literature suggests a few methods, to certain extent, effectively handle low data for better output from the AI models in the context of drug discovery. These are transferring learning (TL), active learning (AL), single or one-shot learning (OSL), multi-task learning (MTL), data augmentation (DA), data synthesis (DS), etc. One different method, which enables sharing of proprietary data on a common platform (without compromising data privacy) to train ML model, is federated learning (FL). In this review, we compare and discuss these methods, their recent applications, and limitations while modeling small molecule data to get the improved output of AI methods in drug discovery. Article also sums up some other novel methods to handle inadequate data.

Towards accurate abdominal tumor segmentation: A 2D model with Position-Aware and Key Slice Feature Sharing.

Abdominal tumor segmentation is a crucial yet challenging step during the screening and diagnosis of tumors. While 3D segmentation models provide powerful performance, they demand substantial computational resources. Additionally, in 3D data, tumors often represent a small portion, leading to imbalanced data and potentially overlooking crucial information. Conversely, 2D segmentation models have a lightweight structure, but disregard the inter-slice correlation, risking the loss of tumor in edge slices. To address these challenges, this paper proposes a novel Position-Aware and Key Slice Feature Sharing 2D tumor segmentation model (PAKS-Net). Leveraging the Swin-Transformer, we effectively model the global features within each slice, facilitating essential information extraction. Furthermore, we introduce a Position-Aware module to capture the spatial relationship between tumors and their corresponding organs, mitigating noise and interference from surrounding organ tissues. To enhance the edge slice segmentation accuracy, we employ key slices to assist in the segmentation of other slices to prioritize tumor regions. Through extensive experiments on three abdominal tumor segmentation CT datasets and a lung tumor segmentation CT dataset, PAKS-Net demonstrates superior performance, reaching 0.893, 0.769, 0.598 and 0.738 tumor DSC on the KiTS19, LiTS17, pancreas and LOTUS datasets, surpassing 3D segmentation models, while remaining computationally efficient with fewer parameters.

Evaluating Virtual Contrast-enhanced MRI (VCE-MRI) in Nasopharyngeal Carcinoma Radiotherapy: A Retrospective Analysis for Primary Gross Tumor Delineation.

PURPOSE: To investigate the potential of virtual contrast-enhanced MRI (VCE-MRI) for gross-tumor-volume (GTV) delineation of nasopharyngeal carcinoma (NPC) using multi-institutional data. METHODS AND MATERIALS: This study retrospectively retrieved T1-weighted (T1w), T2-weighted (T2w) MRI, gadolinium-based contrast-enhanced MRI (CE-MRI) and planning CT of 348 biopsy-proven NPC patients from three oncology centers. A multimodality-guided synergistic neural network (MMgSN-Net) was trained using 288 patients to leverage complementary features in T1w and T2w MRI for VCE-MRI synthesis, which was independently evaluated using 60 patients. Three board-certified radiation oncologists and two medical physicists participated in clinical evaluations in three aspects: image quality assessment of the synthetic VCE-MRI, VCE-MRI in assisting target volume delineation, and effectiveness of VCE-MRI-based contours in treatment planning. The image quality assessment includes distinguishability between VCE-MRI and CE-MRI, clarity of tumor-to-normal tissue interface and veracity of contrast enhancement in tumor invasion risk areas. Primary tumor delineation and treatment planning were manually performed by radiation oncologists and medical physicists, respectively. RESULTS: The mean accuracy to distinguish VCE-MRI from CE-MRI was 31.67%; no significant difference was observed in the clarity of tumor-to-normal tissue interface between VCE-MRI and CE-MRI; for the veracity of contrast enhancement in tumor invasion risk areas, an accuracy of 85.8% was obtained. The image quality assessment results suggest that the image quality of VCE-MRI is highly similar to real CE-MRI. The mean dosimetric difference of planning target volumes were less than 1Gy. CONCLUSIONS: The VCE-MRI is highly promising to replace the use of gadolinium-based CE-MRI in tumor delineation of NPC patients.

Artificial intelligence-assisted decision making: Prediction of optimal level of distal mesorectal margin during transanal total mesorectal excision (taTME) using deep neural network modeling.

BACKGROUND: With steep posterior anorectal angulation, transanal total mesorectal excision (taTME) may have a risk of dissection in the wrong plane or starting higher up, resulting in leaving distal mesorectum behind. Although the distal mesorectal margin can be assessed by preoperative MRI, it needs skilled radiologist and high-definition image for accurate evaluation. This study developed a deep neural network (DNN) to predict the optimal level of distal mesorectal margin. METHODS: A total of 182 pelvic MRI images extracted from the cancer image archive (TCIA) database were included. A DNN was developed using gender, the degree of anterior and posterior anorectal angles as input variables while the difference between anterior and posterior mesorectal distances from anal verge was selected as a target. The predictability power was assessed by regression values (R) which is the correlation between the predicted outputs and actual targets. RESULTS: The anterior angle was an obtuse angle while the posterior angle varied from acute to obtuse with mean angle difference 35.5°±14.6. The mean difference between the anterior and posterior mesorectal end distances was 18.6±6.6mm. The developed DNN had a very close correlation with the target during training, validation, and testing (R=0.99, 0.81, and 0.89, P<0.001). The predicted level of distal mesorectal margin was closely correlated with the actual optimal level (R=0.91, P<0.001). CONCLUSIONS: Artificial intelligence can assist in either making or confirming the preoperative decisions. Furthermore, the developed model can alert the surgeons for this potential risk and the necessity of re-positioning the proctectomy incision.

Transformers for colorectal cancer segmentation in CT imaging.

PURPOSE: Most recently transformer models became the state of the art in various medical image segmentation tasks and challenges, outperforming most of the conventional deep learning approaches. Picking up on that trend, this study aims at applying various transformer models to the highly challenging task of colorectal cancer (CRC) segmentation in CT imaging and assessing how they hold up to the current state-of-the-art convolutional neural network (CNN), the nnUnet. Furthermore, we wanted to investigate the impact of the network size on the resulting accuracies, since transformer models tend to be significantly larger than conventional network architectures. METHODS: For this purpose, six different transformer models, with specific architectural advancements and network sizes were implemented alongside the aforementioned nnUnet and were applied to the CRC segmentation task of the medical segmentation decathlon. RESULTS: The best results were achieved with the Swin-UNETR, D-Former, and VT-Unet, each transformer models, with a Dice similarity coefficient (DSC) of 0.60, 0.59 and 0.59, respectively. Therefore, the current state-of-the-art CNN, the nnUnet could be outperformed by transformer architectures regarding this task. Furthermore, a comparison with the inter-observer variability (IOV) of approx. 0.64 DSC indicates almost expert-level accuracy. The comparatively low IOV emphasizes the complexity and challenge of CRC segmentation, as well as indicating limitations regarding the achievable segmentation accuracy. CONCLUSION: As a result of this study, transformer models underline their current upward trend in producing state-of-the-art results also for the challenging task of CRC segmentation. However, with ever smaller advances in total accuracies, as demonstrated in this study by the on par performances of multiple network variants, other advantages like efficiency, low computation demands, or ease of adaption to new tasks become more and more relevant.

Optimize individualized energy delivery for septic patients using predictive deep learning models.

BACKGROUND AND OBJECTIVES: We aim to establish deep learning models to optimize the individualized energy delivery for septic patients. METHODS AND STUDY DESIGN: We conducted a study of adult septic patients in ICU, collecting 47 indicators for 14 days. We filtered out nutrition-related features and divided the data into datasets according to the three metabolic phases proposed by ESPEN: acute early, acute late, and rehabilitation. We then established optimal energy target models for each phase using deep learning and conducted external validation. RESULTS: A total of 179 patients in training dataset and 98 patients in external validation dataset were included in this study, and total data size was 3115 elements. The age, weight and BMI of the patients were 63.05 (95%CI 60.42-65.68), 61.31(95%CI 59.62-63.00) and 22.70 (95%CI 22.21-23.19), respectively. And 26.0% (72) of the patients were female. The models indicated that the optimal energy targets in the three phases were 900kcal/d, 2300kcal/d, and 2000kcal/d, respectively. Excessive energy intake increased mortality rapidly in the early period of the acute phase. Insufficient energy in the late period of the acute phase significantly raised the mortality as well. For the rehabilitation phase, too much or too little energy delivery were both associated with elevated death risk. CONCLUSIONS: Our study established time-series prediction models for septic patients to optimize energy delivery in the ICU. We recommended permissive underfeeding only in the early acute phase. Later, increased energy intake may improve survival and settle energy debts caused by underfeeding.

A comprehensive dataset for Arabic word sense disambiguation.

This data paper introduces a comprehensive dataset tailored for word sense disambiguation tasks, explicitly focusing on a hundred polysemous words frequently employed in Modern Standard Arabic. The dataset encompasses a diverse set of senses for each word, ranging from 3 to 8, resulting in 367 unique senses. Each word sense is accompanied by contextual sentences comprising ten sentence examples that feature the polysemous word in various contexts. The data collection resulted in a dataset of 3670 samples. Significantly, the dataset is in Arabic, which is known for its rich morphology, complex syntax, and extensive polysemy. The data was meticulously collected from various web sources, spanning news, medicine, finance, and more domains. This inclusivity ensures the dataset's applicability across diverse fields, positioning it as a pivotal resource for Arabic Natural Language Processing (NLP) applications. The data collection timeframe spans from the first of April 2023 to the first of May 2023. The dataset provides comprehensive model learning by including all senses for a frequently used Arabic polysemous term, even rare senses that are infrequently used in real-world contexts, thereby mitigating biases. The dataset comprises synthetic sentences generated by GPT3.5-turbo, addressing instances where rare senses lack sufficient real-world data. The dataset collection process involved initial web scraping, followed by manual sorting to distinguish word senses, supplemented by thorough searches by a human expert to fill in missing contextual sentences. Finally, in instances where online data for rare word senses was lacking or insufficient, synthetic samples were generated. Beyond its primary utility in word sense disambiguation, this dataset holds considerable value for scientists and researchers across various domains, extending its relevance to sentiment analysis applications.

Automatic soft-tissue analysis on orthodontic frontal and lateral facial photographs based on deep learning.

BACKGROUND: To establish the automatic soft-tissue analysis model based on deep learning that performs landmark detection and measurement calculations on orthodontic facial photographs to achieve a more comprehensive quantitative evaluation of soft tissues. METHODS: A total of 578 frontal photographs and 450 lateral photographs of orthodontic patients were collected to construct datasets. All images were manually annotated by two orthodontists with 43 frontal-image landmarks and 17 lateral-image landmarks. Automatic landmark detection models were established, which consisted of a high-resolution network, a feature fusion module based on depthwise separable convolution, and a prediction model based on pixel shuffle. Ten measurements for frontal images and eight measurements for lateral images were defined. Test sets were used to evaluate the model performance, respectively. The mean radial error of landmarks and measurement error were calculated and statistically analysed to evaluate their reliability. RESULTS: The mean radial error was 14.44 ± 17.20 pixels for the landmarks in the frontal images and 13.48 ± 17.12 pixels for the landmarks in the lateral images. There was no statistically significant difference between the model prediction and manual annotation measurements except for the mid facial-lower facial height index. A total of 14 measurements had a high consistency. CONCLUSION: Based on deep learning, we established automatic soft-tissue analysis models for orthodontic facial photographs that can automatically detect 43 frontal-image landmarks and 17 lateral-image landmarks while performing comprehensive soft-tissue measurements. The models can assist orthodontists in efficient and accurate quantitative soft-tissue evaluation for clinical application.

Enhanced deep leaning model for detection and grading of lumbar disc herniation from MRI.

Lumbar disc herniation is one of the most prevalent orthopedic issues in clinical practice. The lumbar spine is a crucial joint for movement and weight-bearing, so back pain can significantly impact the everyday lives of patients and is prone to recurring. The pathogenesis of lumbar disc herniation is complex and diverse, making it difficult to identify and assess after it has occurred. Magnetic resonance imaging (MRI) is the most effective method for detecting injuries, requiring continuous examination by medical experts to determine the extent of the injury. However, the continuous examination process is time-consuming and susceptible to errors. This study proposes an enhanced model, BE-YOLOv5, for hierarchical detection of lumbar disc herniation from MRI images. To tailor the training of the model to the job requirements, a specialized dataset was created. The data was cleaned and improved before the final calibration. A final training set of 2083 data points and a test set of 100 data points were obtained. The YOLOv5 model was enhanced by integrating the attention mechanism module, ECAnet, with a 3 × 3 convolutional kernel size, substituting its feature extraction network with a BiFPN, and implementing structural system pruning. The model achieved an 89.7% mean average precision (mAP) and 48.7 frames per second (FPS) on the test set. In comparison to Faster R-CNN, original YOLOv5, and the latest YOLOv8, this model performs better in terms of both accuracy and speed for the detection and grading of lumbar disc herniation from MRI, validating the effectiveness of multiple enhancement methods. The proposed model is expected to be used for diagnosing lumbar disc herniation from MRI images and to demonstrate efficient and high-precision performance.

Fast, high-quality, and unshielded 0.2 T low-field mobile MRI using minimal hardware resources.

OBJECTIVE: To propose a deep learning-based low-field mobile MRI strategy for fast, high-quality, unshielded imaging using minimal hardware resources. METHODS: Firstly, we analyze the correlation of EMI signals between the sensing coil and the MRI coil to preliminarily verify the feasibility of active EMI shielding using a single sensing coil. Then, a powerful deep learning EMI elimination model is proposed, which can accurately predict the EMI components in the MRI coil signals using EMI signals from at least one sensing coil. Further, deep learning models with different task objectives (super-resolution and denoising) are strategically stacked for multi-level post-processing to enable fast and high-quality low-field MRI. Finally, extensive phantom and brain experiments were conducted on a home-built 0.2 T mobile brain scanner for the evaluation of the proposed strategy. RESULTS: 20 healthy volunteers were recruited to participate in the experiment. The results show that the proposed strategy enables the 0.2 T scanner to generate images with sufficient anatomical information and diagnostic value under unshielded conditions using a single sensing coil. In particular, the EMI elimination outperforms the state-of-the-art deep learning methods and numerical computation methods. In addition, 2 × super-resolution (DDSRNet) and denoising (SwinIR) techniques enable further improvements in imaging speed and quality. DISCUSSION: The proposed strategy enables low-field mobile MRI scanners to achieve fast, high-quality imaging under unshielded conditions using minimal hardware resources, which has great significance for the widespread deployment of low-field mobile MRI scanners.

Training and assessing convolutional neural network performance in automatic vascular segmentation using Ga-68 DOTATATE PET/CT.

To evaluate a convolutional neural network's performance (nnU-Net) in the assessment of vascular contours, calcification and PET tracer activity using Ga-68 DOTATATE PET/CT. Patients who underwent Ga-68 DOTATATE PET/CT imaging over a 12-month period for neuroendocrine investigation were included. Manual cardiac and aortic segmentations were performed by an experienced observer. Scans were randomly allocated in ratio 64:16:20 for training, validation and testing of the nnU-Net model. PET tracer uptake and calcium scoring were compared between segmentation methods and different observers. 116 patients (53.5% female) with a median age of 64.5 years (range 23-79) were included. There were strong, positive correlations between all segmentations (mostly r > 0.98). There were no significant differences between manual and AI segmentation of SUVmean for global cardiac (mean ± SD 0.71 ± 0.22 vs. 0.71 ± 0.22; mean diff 0.001 ± 0.008, p > 0.05), ascending aorta (mean ± SD 0.44 ± 0.14 vs. 0.44 ± 0.14; mean diff 0.002 ± 0.01, p > 0.05), aortic arch (mean ± SD 0.44 ± 0.10 vs. 0.43 ± 0.10; mean diff 0.008 ± 0.16, p > 0.05) and descending aorta (mean ± SD < 0.001; 0.58 ± 0.12 vs. 0.57 ± 0.12; mean diff 0.01 ± 0.03, p > 0.05) contours. There was excellent agreement between the majority of manual and AI segmentation measures (r ≥ 0.80) and in all vascular contour calcium scores. Compared with the manual segmentation approach, the CNN required a significantly lower workflow time. AI segmentation of vascular contours using nnU-Net resulted in very similar measures of PET tracer uptake and vascular calcification when compared to an experienced observer and significantly reduced workflow time.

DSnet: a new dual-branch network for hippocampus subfield segmentation.

The hippocampus is a critical component of the brain and is associated with many neurological disorders. It can be further subdivided into several subfields, and accurate segmentation of these subfields is of great significance for diagnosis and research. However, the structures of hippocampal subfields are irregular and have complex boundaries, and their voxel values are close to surrounding brain tissues, making the segmentation task highly challenging. Currently, many automatic segmentation tools exist for hippocampal subfield segmentation, but they suffer from high time costs and low segmentation accuracy. In this paper, we propose a new dual-branch segmentation network structure (DSnet) based on deep learning for hippocampal subfield segmentation. While traditional convolutional neural network-based methods are effective in capturing hierarchical structures, they struggle to establish long-term dependencies. The DSnet integrates the Transformer architecture and a hybrid attention mechanism, enhancing the network's global perceptual capabilities. Moreover, the dual-branch structure of DSnet leverages the segmentation results of the hippocampal region to facilitate the segmentation of its subfields. We validate the efficacy of our algorithm on the public Kulaga-Yoskovitz dataset. Experimental results indicate that our method is more effective in segmenting hippocampal subfields than conventional single-branch network structures. Compared to the classic 3D U-Net, our proposed DSnet improves the average Dice accuracy of hippocampal subfield segmentation by 0.57%.

Challenges in Reducing Bias Using Post-Processing Fairness for Breast Cancer Stage Classification with Deep Learning.

Breast cancer is the most common cancer affecting women globally. Despite the significant impact of deep learning models on breast cancer diagnosis and treatment, achieving fairness or equitable outcomes across diverse populations remains a challenge when some demographic groups are underrepresented in the training data. We quantified the bias of models trained to predict breast cancer stage from a dataset consisting of 1000 biopsies from 842 patients provided by AIM-Ahead (Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity). Notably, the majority of data (over 70%) were from White patients. We found that prior to post-processing adjustments, all deep learning models we trained consistently performed better for White patients than for non-White patients. After model calibration, we observed mixed results, with only some models demonstrating improved performance. This work provides a case study of bias in breast cancer medical imaging models and highlights the challenges in using post-processing to attempt to achieve fairness.