Biomed Eng Online. 2024 Apr 9;23(1):41. doi: 10.1186/s12938-024-01234-y.

ABSTRACT

BACKGROUND: The timely identification and management of ovarian cancer are critical determinants of patient prognosis. In this study, we developed and validated a deep learning radiomics nomogram (DLR_Nomogram) based on ultrasound (US) imaging to accurately predict the malignant risk of ovarian tumours and compared the diagnostic performance of the DLR_Nomogram to that of the ovarian-adnexal reporting and data system (O-RADS).

METHODS: This study encompasses two research tasks. Patients were randomly divided into training and testing sets in an 8:2 ratio for both tasks. In task 1, we assessed the malignancy risk of 849 patients with ovarian tumours. In task 2, we evaluated the malignancy risk of 391 patients with O-RADS 4 and O-RADS 5 ovarian neoplasms. Three models were developed and validated to predict the risk of malignancy in ovarian tumours. The predicted outcomes of the models for each sample were merged to form a new feature set that was utilised as an input for the logistic regression (LR) model for constructing a combined model, visualised as the DLR_Nomogram. Then, the diagnostic performance of these models was evaluated by the receiver operating characteristic curve (ROC).

RESULTS: The DLR_Nomogram demonstrated superior predictive performance in predicting the malignant risk of ovarian tumours, as evidenced by area under the ROC curve (AUC) values of 0.985 and 0.928 for the training and testing sets of task 1, respectively. The AUC value of its testing set was lower than that of the O-RADS; however, the difference was not statistically significant. The DLR_Nomogram exhibited the highest AUC values of 0.955 and 0.869 in the training and testing sets of task 2, respectively. The DLR_Nomogram showed satisfactory fitting performance for both tasks in Hosmer-Lemeshow testing. Decision curve analysis demonstrated that the DLR_Nomogram yielded greater net clinical benefits for predicting malignant ovarian tumours within a specific range of threshold values.

CONCLUSIONS: The US-based DLR_Nomogram has shown the capability to accurately predict the malignant risk of ovarian tumours, exhibiting a predictive efficacy comparable to that of O-RADS.

PMID:38594729 | DOI:10.1186/s12938-024-01234-y

NPJ Digit Med. 2024 Apr 9;7(1):78. doi: 10.1038/s41746-024-01031-w.

ABSTRACT

The development of diagnostic tools for skin cancer based on artificial intelligence (AI) is increasing rapidly and will likely soon be widely implemented in clinical use. Even though the performance of these algorithms is promising in theory, there is limited evidence on the impact of AI assistance on human diagnostic decisions. Therefore, the aim of this systematic review and meta-analysis was to study the effect of AI assistance on the accuracy of skin cancer diagnosis. We searched PubMed, Embase, IEE Xplore, Scopus and conference proceedings for articles from 1/1/2017 to 11/8/2022. We included studies comparing the performance of clinicians diagnosing at least one skin cancer with and without deep learning-based AI assistance. Summary estimates of sensitivity and specificity of diagnostic accuracy with versus without AI assistance were computed using a bivariate random effects model. We identified 2983 studies, of which ten were eligible for meta-analysis. For clinicians without AI assistance, pooled sensitivity was 74.8% (95% CI 68.6-80.1) and specificity was 81.5% (95% CI 73.9-87.3). For AI-assisted clinicians, the overall sensitivity was 81.1% (95% CI 74.4-86.5) and specificity was 86.1% (95% CI 79.2-90.9). AI benefitted medical professionals of all experience levels in subgroup analyses, with the largest improvement among non-dermatologists. No publication bias was detected, and sensitivity analysis revealed that the findings were robust. AI in the hands of clinicians has the potential to improve diagnostic accuracy in skin cancer diagnosis. Given that most studies were conducted in experimental settings, we encourage future studies to further investigate these potential benefits in real-life settings.

PMID:38594408 | DOI:10.1038/s41746-024-01031-w

Sci Rep. 2024 Apr 9;14(1):8350. doi: 10.1038/s41598-024-58580-z.

ABSTRACT

This study aimed to evaluate the sensitivity of AI in screening acute leukemia and its capability to classify either physiological or pathological cells. Utilizing an acute leukemia orientation tube (ALOT), one of the protocols of Euroflow, flow cytometry efficiently identifies various forms of acute leukemia. However, the analysis of flow cytometry can be time-consuming work. This retrospective study included 241 patients who underwent flow cytometry examination using ALOT between 2017 and 2022. The collected flow cytometry data were used to train an artificial intelligence using deep learning. The trained AI demonstrated a 94.6% sensitivity in detecting acute myeloid leukemia (AML) patients and a 98.2% sensitivity for B-lymphoblastic leukemia (B-ALL) patients. The sensitivities of physiological cells were at least 80%, with variable performance for pathological cells. In conclusion, the AI, trained with ResNet-50 and EverFlow, shows promising results in identifying patients with AML and B-ALL, as well as classifying physiological cells.

PMID:38594383 | DOI:10.1038/s41598-024-58580-z

Mikrochim Acta. 2024 Apr 10;191(5):255. doi: 10.1007/s00604-024-06343-y.

ABSTRACT

Perovskite quantum dots (PQDs) are novel nanomaterials wherein perovskites are used to formulate quantum dots (QDs). The present study utilizes the excellent fluorescence quantum yields of these nanomaterials to detect 16S rRNA of circulating microbiome for risk assessment of cardiovascular diseases (CVDs). A long short-term memory (LSTM) deep learning model was used to find the association of the circulating bacterial species with CVD risk, which showed the abundance of three different bacterial species (Bauldia litoralis (BL), Hymenobacter properus (HYM), and Virgisporangium myanmarense (VIG)). The observations suggested that the developed nano-sensor provides high sensitivity, selectivity, and applicability. The observed sensitivities for Bauldia litoralis, Hymenobacter properus, and Virgisporangium myanmarense were 0.606, 0.300, and 0.281 fg, respectively. The developed sensor eliminates the need for labelling, amplification, quantification, and biochemical assessments, which are more labour-intensive, time-consuming, and less reliable. Due to the rapid detection time, user-friendly nature, and stability, the proposed method has a significant advantage in facilitating point-of-care testing of CVDs in the future. This may also facilitate easy integration of the approach into various healthcare settings, making it accessible and valuable for resource-constrained environments.

PMID:38594377 | DOI:10.1007/s00604-024-06343-y

NPJ Precis Oncol. 2024 Apr 9;8(1):88. doi: 10.1038/s41698-024-00560-7.

ABSTRACT

Microsatellite instability-high (MSI-H) is a tumor-agnostic biomarker for immune checkpoint inhibitor therapy. However, MSI status is not routinely tested in prostate cancer, in part due to low prevalence and assay cost. As such, prediction of MSI status from hematoxylin and eosin (H&E) stained whole-slide images (WSIs) could identify prostate cancer patients most likely to benefit from confirmatory testing to evaluate their eligibility for immunotherapy and need for Lynch syndrome testing. Prostate biopsies and surgical resections from prostate cancer patients referred to our institution were analyzed. MSI status was determined by next-generation sequencing. Patients sequenced before a cutoff date formed an algorithm development set (n = 4015, MSI-H 1.8%) and a paired validation set (n = 173, MSI-H 19.7%) that consisted of two serial sections from each sample, one stained and scanned internally and the other at an external site. Patients sequenced after the cutoff date formed a temporally independent validation set (n = 1350, MSI-H 2.3%). Attention-based multiple instance learning models were trained to predict MSI-H from H&E WSIs. The predictor achieved area under the receiver operating characteristic curve values of 0.78 (95% CI [0.69-0.86]), 0.72 (95% CI [0.63-0.81]), and 0.72 (95% CI [0.62-0.82]) on the internally prepared, externally prepared, and temporal validation sets, respectively, showing effective predictability and generalization to both external staining/scanning processes and temporally independent samples. While MSI-H status is significantly correlated with Gleason score, the model remained predictive within each Gleason score subgroup.

PMID:38594360 | DOI:10.1038/s41698-024-00560-7

NPJ Precis Oncol. 2024 Apr 9;8(1):89. doi: 10.1038/s41698-024-00580-3.

ABSTRACT

The development of deep learning (DL) models to predict the consensus molecular subtypes (CMS) from histopathology images (imCMS) is a promising and cost-effective strategy to support patient stratification. Here, we investigate whether imCMS calls generated from whole slide histopathology images (WSIs) of rectal cancer (RC) pre-treatment biopsies are associated with pathological complete response (pCR) to neoadjuvant long course chemoradiotherapy (LCRT) with single agent fluoropyrimidine. DL models were trained to classify WSIs of colorectal cancers stained with hematoxylin and eosin into one of the four CMS classes using a multi-centric dataset of resection and biopsy specimens (n = 1057 WSIs) with paired transcriptional data. Classifiers were tested on a held out RC biopsy cohort (ARISTOTLE) and correlated with pCR to LCRT in an independent dataset merging two RC cohorts (ARISTOTLE, n = 114 and SALZBURG, n = 55 patients). DL models predicted CMS with high classification performance in multiple comparative analyses. In the independent cohorts (ARISTOTLE, SALZBURG), cases with WSIs classified as imCMS1 had a significantly higher likelihood of achieving pCR (OR = 2.69, 95% CI 1.01-7.17, p = 0.048). Conversely, imCMS4 was associated with lack of pCR (OR = 0.25, 95% CI 0.07-0.88, p = 0.031). Classification maps demonstrated pathologist-interpretable associations with high stromal content in imCMS4 cases, associated with poor outcome. No significant association was found in imCMS2 or imCMS3. imCMS classification of pre-treatment biopsies is a fast and inexpensive solution to identify patient groups that could benefit from neoadjuvant LCRT. The significant associations between imCMS1/imCMS4 with pCR suggest the existence of predictive morphological features that could enhance standard pathological assessment.

PMID:38594327 | DOI:10.1038/s41698-024-00580-3

J Med Imaging Radiat Sci. 2024 Apr 8:S1939-8654(24)00100-0. doi: 10.1016/j.jmir.2024.03.046. Online ahead of print.

ABSTRACT

INTRODUCTION: Radiologists have extensively employed the interpretation of chest X-rays (CXR) to identify visual markers indicative of COVID-19 infection, offering an alternative approach for the screening of infected individuals. This research article presents CovMediScanX, a deep learning-based framework designed for a rapid and automated diagnosis of COVID-19 from CXR scan images.

METHODS: The proposed approach encompasses gathering and preprocessing CXR image datasets, training deep learning-based custom-made Convolutional Neural Network (CNN), pre-trained and hybrid transfer learning models, identifying the highest-performing model based on key evaluation metrics, and embedding this model into a web interface called CovMediScanX, designed for radiologists to detect the COVID-19 status in new CXR images.

RESULTS: The custom-made CNN model obtained a remarkable testing accuracy of 94.32% outperforming other models. CovMediScanX, employing the custom-made CNN underwent evaluation with an independent dataset also. The images in the independent dataset are sourced from a scanning machine that is entirely different from those used for the training dataset, highlighting a clear distinction of datasets in their origins. The evaluation outcome highlighted the framework's capability to accurately detect COVID-19 cases, showcasing encouraging results with a precision of 73% and a recall of 84% for positive cases. However, the model requires further enhancement, particularly in improving its detection of normal cases, as evidenced by lower precision and recall rates.

CONCLUSION: The research proposes CovMediScanX framework that demonstrates promising potential in automatically identifying COVID-19 cases from CXR images. While the model's overall performance on independent data needs improvement, it is evident that addressing bias through the inclusion of diverse data sources during training could further enhance accuracy and reliability.

PMID:38594085 | DOI:10.1016/j.jmir.2024.03.046

J Am Coll Cardiol. 2024 Mar 26:S0735-1097(24)06742-1. doi: 10.1016/j.jacc.2024.03.400. Online ahead of print.

ABSTRACT

Recent AI advancements in cardiovascular care offer potential enhancements in diagnosis, treatment, and outcomes. Innovations to date focus on automating measurements, enhancing image quality, and detecting diseases using novel methods. Applications span wearables, electrocardiograms, echocardiography, angiography, genetics, and more. AI models detect diseases from electrocardiograms at accuracy not previously achieved by technology or human experts, including reduced ejection fraction, valvular heart disease, and other cardiomyopathies. However, AI's unique characteristics necessitates rigorous validation by addressing training methods, real-world efficacy, equity concerns, and long-term reliability. Despite an exponentially growing number of studies in cardiovascular AI, trials showing improvement in outcomes remain lacking. A number are currently underway. Embracing this rapidly evolving technology while setting a high evaluation benchmark will be crucial for cardiology to leverage AI to enhance patient care and the provider experience.

PMID:38593946 | DOI:10.1016/j.jacc.2024.03.400

J Am Coll Cardiol. 2024 Mar 26:S0735-1097(24)06744-5. doi: 10.1016/j.jacc.2024.03.401. Online ahead of print.

ABSTRACT

Recent Artificial Intelligence (AI) advancements in cardiovascular care offer potential enhancements in effective diagnosis, treatment, and outcomes. Over 600 Food and Drug Administration (FDA)-approved clinical AI algorithms now exist, with 10% focusing on cardiovascular applications, highlighting the growing opportunities for AI to augment care. This review discusses the latest advancements in the field of AI, with a particular focus on the utilization of multimodal inputs and the field of generative AI. Further discussions in this review involve an approach to understanding the larger context in which AI-augmented care may exist, and include a discussion of the need for rigorous evaluation, appropriate infrastructure for deployment, ethics and equity assessments, regulatory oversight, and viable business cases for deployment. Embracing this rapidly evolving technology while setting an appropriately high evaluation benchmark with careful and patient-centered implementation will be crucial for cardiology to leverage AI to enhance patient care and the provider experience.

PMID:38593945 | DOI:10.1016/j.jacc.2024.03.401

Sci Total Environ. 2024 Apr 7:172246. doi: 10.1016/j.scitotenv.2024.172246. Online ahead of print.

ABSTRACT

Proactive management of pumping stations using artificial intelligence (AI) technology is vital for effectively mitigating the impacts of flood events caused by climate change. Accurate water level forecasts are pivotal in advancing the intelligent operation of pumping stations. This study proposed a novel Transformer-LSTM model to offer accurate multi-step-ahead forecasts of the flood storage pond (FSP) and river water levels for the Zhongshan pumping station in Taipei, Taiwan. A total of 19,647 ten-minute-based datasets of pumping operation and storm sewer, FSP, and river water levels were collected between 2014 and 2020 and further divided into training (70 %), validation (10 %), and test (20 %) datasets for model construction. The results demonstrate that the proposed model dramatically outperforms benchmark models by producing more accurate and reliable water level forecasts at 10-minute (T + 1) to 60-minute (T + 6) horizons. The proposed model effectively enhances the connections between input factors through the Transformer module and increases the connectivity across consecutive time series using the LSTM module. This study reveals interconnected dynamics among pumping operation and storm sewer, FSP, and river water levels, enhancing flood management. Understanding these dynamics is crucial for effective execution of management strategies and infrastructure revitalization against climate impacts. The Transformer-LSTM model's forecasts encourage water practices, resilience, and disaster risk reduction for extreme weather events.

PMID:38593878 | DOI:10.1016/j.scitotenv.2024.172246

J Endocrinol. 2024 Apr 1:JOE-23-0204. doi: 10.1530/JOE-23-0204. Online ahead of print.

ABSTRACT

The mouse estrous cycle is divided into four stages: proestrus (P), estrus (E), metestrus (M) and diestrus (D). The estrous cycle affects reproductive hormone levels in a wide variety of tissues. Therefore, to obtain reliable results from female mice, it is important to know the estrous cycle stage during sampling. The stage can be analyzed from a vaginal smear under a microscope. However, it is time-consuming, and the results vary between evaluators. Here, we present an accurate and reproducible method for staging the mouse estrous cycle in digital whole slide images (WSIs) of vaginal smears. We developed a model using a deep convolutional neural network (CNN) in a cloud-based platform, Aiforia Create. The CNN was trained by supervised pixel-level multiclass semantic segmentation of image features from 171 hematoxylin-stained samples. The model was validated by comparing the results obtained by CNN with those of four independent researchers. The validation data included three separate studies comprising altogether 148 slides. The total agreement attested by the Fleiss kappa value between the validators and the CNN was excellent (0.75), and when D, E and P were analyzed separately, the kappa values were 0.89, 0.79 and 0.74, respectively. The M stage is short and not well defined by the researchers. Thus, identification of the M stage by the CNN was challenging due to the lack of proper ground truth, and the kappa value was 0.26. We conclude that our model is reliable and effective for classifying the estrous cycle stages in female mice.

PMID:38593833 | DOI:10.1530/JOE-23-0204

Phys Med Biol. 2024 Apr 9. doi: 10.1088/1361-6560/ad3cb2. Online ahead of print.

ABSTRACT

To go beyond the deficiencies of the three conventional multimodal fusion strategies (i.e., input-, feature- and output-level fusion), we propose a bidirectional attention-aware fluid pyramid feature integrated fusion network (BAF-Net) with cross-modal interactions for multimodal medical image diagnosis and prognosis.&#xD;Approach: BAF-Net is composed of two identical branches to preserve the unimodal features and one bidirectional attention-aware distillation stream to progressively assimilate cross-modal complements and to learn supplementary features in both bottom-up and top-down processes. Fluid pyramid connections were adopted to integrate the hierarchical features at different levels of the network, and channel-wise attention modules were exploited to mitigate cross-modal cross-level incompatibility. Furthermore, depth-wise separable convolution was introduced to fuse the cross-modal cross-level features to alleviate the increase in parameters to a great extent. The generalization abilities of BAF-Net were evaluated in terms of two clinical tasks: (1) An in-house PET-CT dataset with 174 patients for differentiation between lung cancer and pulmonary tuberculosis. (2) A public multicenter PET-CT head and neck cancer dataset with 800 patients from nine centers for overall survival prediction.&#xD;Main results: On the LC-PTB dataset, improved performance was found in BAF-Net (AUC = 0.7342) compared with input-level fusion model (AUC = 0.6825; p < 0.05), feature-level fusion model (AUC = 0.6968; p = 0.0547), output-level fusion model (AUC = 0.7011; p < 0.05). On the H&N cancer dataset, BAF-Net (C-index = 0.7241) outperformed the input-, feature-, and output-level fusion model, with 2.95%, 3.77%, and 1.52% increments of C-index (p = 0.3336, 0.0479 and 0.2911, respectively). The ablation experiments demonstrated the effectiveness of all the designed modules regarding all the evaluated metrics in both datasets.&#xD;Significance: Extensive experiments on two datasets demonstrated better performance and robustness of BAF-Net than three conventional fusion strategies and PET or CT unimodal network in terms of diagnosis and prognosis.

PMID:38593831 | DOI:10.1088/1361-6560/ad3cb2

Phys Med Biol. 2024 Apr 9. doi: 10.1088/1361-6560/ad3cb3. Online ahead of print.

ABSTRACT

OBJECTIVE: Automatic medical image segmentation is crucial for accurately isolating target tissue areas in the image from background tissues, facilitating precise diagnoses and procedures. While the proliferation of publicly available clinical datasets led to the development of deep learning-based medical image segmentation methods, a generalized, accurate, robust, and reliable approach across diverse imaging modalities remains elusive.

APPROACH: This paper proposes a novel high-resolution parallel generative adversarial network (pGAN)-based generalized deep learning method for automatic segmentation of medical images from diverse imaging modalities. The proposed method showcases better performance and generalizability by incorporating novel components such as partial hybrid transfer learning, discrete wavelet transform (DWT)-based multilayer and multiresolution feature fusion in the encoder, and a dual mode attention gate in the decoder of the multi-resolution U-Net-based GAN. With multi-objective adversarial training loss functions including a unique reciprocal loss for enforcing cooperative learning in pGANs, it further enhances the robustness and accuracy of the segmentation map.

MAIN RESULTS: Experimental evaluations conducted on nine diverse publicly available medical image segmentation datasets, including PhysioNet ICH, BUSI, CVC-ClinicDB, MoNuSeg, GLAS, ISIC-2018, DRIVE, Montgomery, and PROMISE12, demonstrate the proposed method's superior performance. The proposed method achieves mean F1 scores of 79.53%, 88.68%, 82.50%, 93.25%, 90.40%, 94.19%, 81.65%, 98.48%, and 90.79%, respectively, on the above datasets, surpass state-of-the-art segmentation methods. Furthermore, our proposed method demonstrates robust multi-domain segmentation capabilities, exhibiting consistent and reliable performance. The assessment of the model's proficiency in accurately identifying small details indicates that the high-resolution generalized medical image segmentation network (Hi-gMISnet) is more precise in segmenting even when the target area is very small.

SIGNIFICANCE: The proposed method provides robust and reliable segmentation performance on medical images, and thus it has the potential to be used in a clinical setting for the diagnosis of patients.

PMID:38593830 | DOI:10.1088/1361-6560/ad3cb3

Phys Med Biol. 2024 Apr 9. doi: 10.1088/1361-6560/ad3cb1. Online ahead of print.

ABSTRACT

&#xD;Radiomics and deep learning techniques have become integral in meningioma grading. The combination of these approaches holds the potential to enhance classification accuracy. Given the frequent occurrence of peritumoral edema (PTE) in meningiomas, investigating the potential value of PTE requires further research.&#xD;Objectives:&#xD;To address the challenge of meningioma grading, this study introduces a unique approach that integrates radiomics and deep learning techniques. The primary focus is on the development of a Transfer Learning-based Meningioma Feature Extraction Model (MFEM), leveraging both Vision Transformer (ViT) and Convolutional Neural Network (CNN) architectures. Furthermore, the study explores the potential significance of the peritumoral edema (PTE) region in enhancing the grading process.&#xD;Materials and Methods:&#xD;A retrospective study was conducted involving 98 meningioma patients, with 60 classified as low-grade meningiomas and 38 as high-grade meningiomas. PTE was observed in 51.02% of low-grade meningiomas patients and 89.47% of high-grade meningiomas patients. Magnetic resonance images were acquired using a GE Signa HDxt 1.5T MRI scanner, incorporating T2-weighted Fluid-Attenuated Inversion Recovery (T2 Flair) sequences. A Transfer Learning-based Meningioma Feature Extraction Model (MFEM) was constructed by combining ViT and CNN models to extract deep features from the Transformer layer, alongside radiomics features. These were then utilized as input for a machine learning classifier to accurately grade meningiomas.&#xD;Results:&#xD;The proposed method demonstrated excellent grading accuracy and robustness on the meningioma dataset, offering valuable guidance for treatment decisions. The approach achieved 92.86% accuracy, 93.44% precision, 95% sensitivity, and 89.47% specificity.&#xD;Conclusion:&#xD;The radiomics and deep learning-based approach presented in this study offers a reliable method for preoperative meningioma grading. This innovative method not only enhances accuracy but also reduces observer subjectivity, thereby contributing to improved clinical decision-making.&#xD.

PMID:38593827 | DOI:10.1088/1361-6560/ad3cb1

Phys Med Biol. 2024 Apr 9. doi: 10.1088/1361-6560/ad3c91. Online ahead of print.

ABSTRACT

OBJECTIVE: The textures and detailed structures in computed tomography (CT) images are highly desirable for clinical diagnosis. This study aims to expand the current body of work on textures and details preserving convolutional neural networks for low-dose CT (LDCT) image denoising task.

APPROACH: This study proposed a novel Multi-scale Feature Aggregation and Fusion network (MFAF-net) for LDCT image denoising. Specifically, we proposed a Multi-scale Residual Feature Aggregation Module (MRFAM) to characterize multi-scale structural information in CT images, which captures regional-specific inter-scale variations using learned weights. We further proposed a Cross-level Feature Fusion Module (CFFM) to integrate cross-level features, which adaptively weights the contributions of features from encoder to decoder by using a Spatial Pyramid Attention (SPA) mechanism. Moreover, we proposed a Self-supervised Multi-level Perceptual Loss Module (SMPLM) to generate multi-level auxiliary perceptual supervision for recovery of salient textures and structures of tissues and lesions in CT images, which takes advantage of abundant semantic information at various levels. We introduced parameters for the perceptual loss to adaptively weight the contributions of auxiliary features of different levels and we also introduced an automatic parameter tuning strategy for these parameters.

MAIN RESULTS: Extensive experimental studies were performed to validate the effectiveness of the proposed method. Experimental results demonstrate that the proposed method can achieve better performance on both fine textures preservation and noise suppression for CT image denoising task compared with other competitive CNN based methods.

SIGNIFICANCE: The proposed MFAF-net takes advantage of multi-scale receptive fields, cross-level features integration and self-supervised multi-level perceptual loss, enabling more effective recovering of fine textures and detailed structures of tissues and lesions in CT images.

PMID:38593821 | DOI:10.1088/1361-6560/ad3c91

Phys Med Biol. 2024 Apr 9. doi: 10.1088/1361-6560/ad3c8e. Online ahead of print.

ABSTRACT

Limited-angle computed tomography (CT) presents a challenge due to its ill-posed nature. In such scenarios, analytical reconstruction methods often exhibit severe artifacts. To tackle this inverse problem, several supervised deep learning-based approaches have been proposed. However, they are constrained by limitations such as generalization issue and the difficulty of acquiring a large amount of paired CT images.&#xD;Approach. In this work, we propose an iterative neural reconstruction framework designed for limited-angle CT. By leveraging a coordinate-based neural representation, we formulate tomographic reconstruction as a convex optimization problem involving a deep neural network. We then employ differentiable projection layer to optimize this network by minimizing the discrepancy between the predicted and measured projection data. In addition, we introduce a prior-based weight initialization method to ensure the network starts optimization with an informed initial guess. This strategic initialization significantly improves the quality of iterative reconstruction by stabilizing the divergent behavior in ill-posed neural fields. Our method operates in a self-supervised manner, thereby eliminating the need for extensive data.&#xD;Main results. The proposed method outperforms other iterative and learning-based methods. Experimental results on XCAT and Mayo Clinic datasets demonstrate the effectiveness of our approach in restoring anatomical features as well as structures. This finding was substantiated by visual inspections and quantitative evaluations using NRMSE, PSNR, and SSIM. Moreover, we conduct a comprehensive investigation into the divergent behavior of iterative neural reconstruction, thus revealing its suboptimal convergence when starting from scratch. In contrast, our method consistently produced accurate images by incorporating an initial estimate as informed initialization.&#xD;Significance. This work showcases the feasibility to reconstruct high-fidelity CT images from limited-angle X-ray projections. The proposed methodology introduces a novel data-free approach to enhance medical imaging, holding promise across various clinical applications.

PMID:38593820 | DOI:10.1088/1361-6560/ad3c8e

Artif Intell Med. 2024 Apr 4;151:102866. doi: 10.1016/j.artmed.2024.102866. Online ahead of print.

ABSTRACT

An echocardiogram is a sophisticated ultrasound imaging technique employed to diagnose heart conditions. The transthoracic echocardiogram, one of the most prevalent types, is instrumental in evaluating significant cardiac diseases. However, interpreting its results heavily relies on the clinician's expertise. In this context, artificial intelligence has emerged as a vital tool for helping clinicians. This study critically analyzes key state-of-the-art research that uses deep learning techniques to automate transthoracic echocardiogram analysis and support clinical judgments. We have systematically organized and categorized articles that proffer solutions for view classification, enhancement of image quality and dataset, segmentation and identification of cardiac structures, detection of cardiac function abnormalities, and quantification of cardiac functions. We compared the performance of various deep learning approaches within each category, identifying the most promising methods. Additionally, we highlight limitations in current research and explore promising avenues for future exploration. These include addressing generalizability issues, incorporating novel AI approaches, and tackling the analysis of rare cardiac diseases.

PMID:38593684 | DOI:10.1016/j.artmed.2024.102866

JMIR Ment Health. 2024 Apr 9;11:e55988. doi: 10.2196/55988.

ABSTRACT

BACKGROUND: Large language models (LLMs) hold potential for mental health applications. However, their opaque alignment processes may embed biases that shape problematic perspectives. Evaluating the values embedded within LLMs that guide their decision-making have ethical importance. Schwartz's theory of basic values (STBV) provides a framework for quantifying cultural value orientations and has shown utility for examining values in mental health contexts, including cultural, diagnostic, and therapist-client dynamics.

OBJECTIVE: This study aimed to (1) evaluate whether the STBV can measure value-like constructs within leading LLMs and (2) determine whether LLMs exhibit distinct value-like patterns from humans and each other.

METHODS: In total, 4 LLMs (Bard, Claude 2, Generative Pretrained Transformer [GPT]-3.5, GPT-4) were anthropomorphized and instructed to complete the Portrait Values Questionnaire-Revised (PVQ-RR) to assess value-like constructs. Their responses over 10 trials were analyzed for reliability and validity. To benchmark the LLMs' value profiles, their results were compared to published data from a diverse sample of 53,472 individuals across 49 nations who had completed the PVQ-RR. This allowed us to assess whether the LLMs diverged from established human value patterns across cultural groups. Value profiles were also compared between models via statistical tests.

RESULTS: The PVQ-RR showed good reliability and validity for quantifying value-like infrastructure within the LLMs. However, substantial divergence emerged between the LLMs' value profiles and population data. The models lacked consensus and exhibited distinct motivational biases, reflecting opaque alignment processes. For example, all models prioritized universalism and self-direction, while de-emphasizing achievement, power, and security relative to humans. Successful discriminant analysis differentiated the 4 LLMs' distinct value profiles. Further examination found the biased value profiles strongly predicted the LLMs' responses when presented with mental health dilemmas requiring choosing between opposing values. This provided further validation for the models embedding distinct motivational value-like constructs that shape their decision-making.

CONCLUSIONS: This study leveraged the STBV to map the motivational value-like infrastructure underpinning leading LLMs. Although the study demonstrated the STBV can effectively characterize value-like infrastructure within LLMs, substantial divergence from human values raises ethical concerns about aligning these models with mental health applications. The biases toward certain cultural value sets pose risks if integrated without proper safeguards. For example, prioritizing universalism could promote unconditional acceptance even when clinically unwise. Furthermore, the differences between the LLMs underscore the need to standardize alignment processes to capture true cultural diversity. Thus, any responsible integration of LLMs into mental health care must account for their embedded biases and motivation mismatches to ensure equitable delivery across diverse populations. Achieving this will require transparency and refinement of alignment techniques to instill comprehensive human values.

PMID:38593424 | DOI:10.2196/55988

J Refract Surg. 2024 Apr;40(4):e199-e207. doi: 10.3928/1081597X-20240214-01. Epub 2024 Apr 1.

ABSTRACT

PURPOSE: To investigate the efficacy of incorporating Generative Adversarial Network (GAN) and synthetic images in enhancing the performance of a convolutional neural network (CNN) for automated estimation of Implantable Collamer Lens (ICL) vault using anterior segment optical coherence tomography (AS-OCT).

METHODS: This study was a retrospective evaluation using synthetic data and real patient images in a deep learning framework. Synthetic ICL AS-OCT scans were generated using GANs and a secondary image editing algorithm, creating approximately 100,000 synthetic images. These were used alongside real patient scans to train a CNN for estimating ICL vault distance. The model's performance was evaluated using statistical metrics such as mean absolute percentage error (MAPE), mean absolute error (MAE), root mean squared error (RMSE), and coefficient of determination (R2) for the estimation of ICL vault distance.

RESULTS: The study analyzed 4,557 AS-OCT B-scans from 138 eyes of 103 patients for training. An independent, retrospectively collected dataset of 2,454 AS-OCT images from 88 eyes of 56 patients, used prospectively for evaluation, served as the test set. When trained solely on real images, the CNN achieved a MAPE of 15.31%, MAE of 44.68 µm, and RMSE of 63.3 µm. However, with the inclusion of GAN-generated and algorithmically edited synthetic images, the performance significantly improved, achieving a MAPE of 8.09%, MAE of 24.83 µm, and RMSE of 32.26 µm. The R2 value was +0.98, indicating a strong positive correlation between actual and predicted ICL vault distances (P < .01). No statistically significant difference was observed between measured and predicted vault values (P = .58).

CONCLUSIONS: The integration of GAN-generated and edited synthetic images substantially enhanced ICL vault estimation, demonstrating the efficacy of GANs and synthetic data in enhancing OCT image analysis accuracy. This model not only shows potential for assisting postoperative ICL evaluations, but also for improving OCT automation when data paucity is an issue. [J Refract Surg. 2024;40(4):e199-e207.].

PMID:38593258 | DOI:10.3928/1081597X-20240214-01

PLoS One. 2024 Apr 9;19(4):e0300640. doi: 10.1371/journal.pone.0300640. eCollection 2024.

ABSTRACT

Traffic accidents remain a leading cause of fatalities, injuries, and significant disruptions on highways. Comprehending the contributing factors to these occurrences is paramount in enhancing safety on road networks. Recent studies have demonstrated the utility of predictive modeling in gaining insights into the factors that precipitate accidents. However, there has been a dearth of focus on explaining the inner workings of complex machine learning and deep learning models and the manner in which various features influence accident prediction models. As a result, there is a risk that these models may be seen as black boxes, and their findings may not be fully trusted by stakeholders. The main objective of this study is to create predictive models using various transfer learning techniques and to provide insights into the most impactful factors using Shapley values. To predict the severity of injuries in accidents, Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Residual Networks (ResNet), EfficientNetB4, InceptionV3, Extreme Inception (Xception), and MobileNet are employed. Among the models, the MobileNet showed the highest results with 98.17% accuracy. Additionally, by understanding how different features affect accident prediction models, researchers can gain a deeper understanding of the factors that contribute to accidents and develop more effective interventions to prevent them.

PMID:38593130 | DOI:10.1371/journal.pone.0300640