Comput Methods Programs Biomed. 2024 Jan 17;245:108037. doi: 10.1016/j.cmpb.2024.108037. Online ahead of print.

ABSTRACT

BACKGROUND: aortic stenosis is a common heart valve disease that mainly affects older people in developed countries. Its early detection is crucial to prevent the irreversible disease progression and, eventually, death. A typical screening technique to detect stenosis uses echocardiograms; however, variations introduced by other tissues, camera movements, and uneven lighting can hamper the visual inspection, leading to misdiagnosis. To address these issues, effective solutions involve employing deep learning algorithms to assist clinicians in detecting and classifying stenosis by developing models that can predict this pathology from single heart views. Although promising, the visual information conveyed by a single image may not be sufficient for an accurate diagnosis, especially when using an automatic system; thus, this indicates that different solutions should be explored.

METHODOLOGY: following this rationale, this paper proposes a novel deep learning architecture, composed of a multi-view, multi-scale feature extractor, and a transformer encoder (MV-MS-FETE) to predict stenosis from parasternal long and short-axis views. In particular, starting from the latter, the designed model extracts relevant features at multiple scales along its feature extractor component and takes advantage of a transformer encoder to perform the final classification.

RESULTS: experiments were performed on the recently released Tufts medical echocardiogram public dataset, which comprises 27,788 images split into training, validation, and test sets. Due to the recent release of this collection, tests were also conducted on several state-of-the-art models to create multi-view and single-view benchmarks. For all models, standard classification metrics were computed (e.g., precision, F1-score). The obtained results show that the proposed approach outperforms other multi-view methods in terms of accuracy and F1-score and has more stable performance throughout the training procedure. Furthermore, the experiments also highlight that multi-view methods generally perform better than their single-view counterparts.

CONCLUSION: this paper introduces a novel multi-view and multi-scale model for aortic stenosis recognition, as well as three benchmarks to evaluate it, effectively providing multi-view and single-view comparisons that fully highlight the model's effectiveness in aiding clinicians in performing diagnoses while also producing several baselines for the aortic stenosis recognition task.

PMID:38271793 | DOI:10.1016/j.cmpb.2024.108037

Comput Methods Programs Biomed. 2024 Jan 18;245:108038. doi: 10.1016/j.cmpb.2024.108038. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: Image segmentation is an essential component in medical image analysis. The case of 3D images such as MRI is particularly challenging and time consuming. Interactive or semi-automatic methods are thus highly desirable. However, existing methods do not exploit the typical sequentiality of real user interactions. This is due to the interaction memory used in these systems, which discards ordering. In contrast, we argue that the order of the user corrections should be used for training and lead to performance improvements.

METHODS: We contribute to solving this problem by proposing a general multi-class deep learning-based interactive framework for image segmentation, which embeds a base network in a user interaction loop with a user feedback memory. We propose to model the memory explicitly as a sequence of consecutive system states, from which the features can be learned, generally learning from the segmentation refinement process. Training is a major difficulty owing to the network's input being dependent on the previous output. We adapt the network to this loop by introducing a virtual user in the training process, modelled by dynamically simulating the iterative user feedback.

RESULTS: We evaluated our framework against existing methods on the complex task of multi-class semantic instance female pelvis MRI segmentation with 5 classes, including up to 27 tumour instances, using a segmentation dataset collected in our hospital, and on liver and pancreas CT segmentation, using public datasets. We conducted a user evaluation, involving both senior and junior medical personnel in matching and adjacent areas of expertise. We observed an annotation time reduction with 5'56" for our framework against 25' on average for classical tools. We systematically evaluated the influence of the number of clicks on the segmentation accuracy. A single interaction round our framework outperforms existing automatic systems with a comparable setup. We provide an ablation study and show that our framework outperforms existing interactive systems.

CONCLUSIONS: Our framework largely outperforms existing systems in accuracy, with the largest impact on the smallest, most difficult classes, and drastically reduces the average user segmentation time with fast inference at 47.2±6.2 ms per image.

PMID:38271792 | DOI:10.1016/j.cmpb.2024.108038

Biomark Res. 2024 Jan 25;12(1):12. doi: 10.1186/s40364-024-00561-5.

ABSTRACT

BACKGROUND: Accurate prediction of tumor molecular alterations is vital for optimizing cancer treatment. Traditional tissue-based approaches encounter limitations due to invasiveness, heterogeneity, and molecular dynamic changes. We aim to develop and validate a deep learning radiomics framework to obtain imaging features that reflect various molecular changes, aiding first-line treatment decisions for cancer patients.

METHODS: We conducted a retrospective study involving 508 NSCLC patients from three institutions, incorporating CT images and clinicopathologic data. Two radiomic scores and a deep network feature were constructed on three data sources in the 3D tumor region. Using these features, we developed and validated the 'Deep-RadScore,' a deep learning radiomics model to predict prognostic factors, gene mutations, and immune molecule expression levels.

FINDINGS: The Deep-RadScore exhibits strong discrimination for tumor molecular features. In the independent test cohort, it achieved impressive AUCs: 0.889 for lymphovascular invasion, 0.903 for pleural invasion, 0.894 for T staging; 0.884 for EGFR and ALK, 0.896 for KRAS and PIK3CA, 0.889 for TP53, 0.895 for ROS1; and 0.893 for PD-1/PD-L1. Fusing features yielded optimal predictive power, surpassing any single imaging feature. Correlation and interpretability analyses confirmed the effectiveness of customized deep network features in capturing additional imaging phenotypes beyond known radiomic features.

INTERPRETATION: This proof-of-concept framework demonstrates that new biomarkers across imaging features and molecular phenotypes can be provided by fusing radiomic features and deep network features from multiple data sources. This holds the potential to offer valuable insights for radiological phenotyping in characterizing diverse tumor molecular alterations, thereby advancing the pursuit of non-invasive personalized treatment for NSCLC patients.

PMID:38273398 | DOI:10.1186/s40364-024-00561-5

BMC Med Inform Decis Mak. 2024 Jan 26;24(1):25. doi: 10.1186/s12911-024-02431-4.

ABSTRACT

BACKGROUND: The epiretinal membrane (ERM) is a common retinal disorder characterized by abnormal fibrocellular tissue at the vitreomacular interface. Most patients with ERM are asymptomatic at early stages. Therefore, screening for ERM will become increasingly important. Despite the high prevalence of ERM, few deep learning studies have investigated ERM detection in the color fundus photography (CFP) domain. In this study, we built a generative model to enhance ERM detection performance in the CFP.

METHODS: This deep learning study retrospectively collected 302 ERM and 1,250 healthy CFP data points from a healthcare center. The generative model using StyleGAN2 was trained using single-center data. EfficientNetB0 with StyleGAN2-based augmentation was validated using independent internal single-center data and external datasets. We randomly assigned healthcare center data to the development (80%) and internal validation (20%) datasets. Data from two publicly accessible sources were used as external validation datasets.

RESULTS: StyleGAN2 facilitated realistic CFP synthesis with the characteristic cellophane reflex features of the ERM. The proposed method with StyleGAN2-based augmentation outperformed the typical transfer learning without a generative adversarial network. The proposed model achieved an area under the receiver operating characteristic (AUC) curve of 0.926 for internal validation. AUCs of 0.951 and 0.914 were obtained for the two external validation datasets. Compared with the deep learning model without augmentation, StyleGAN2-based augmentation improved the detection performance and contributed to the focus on the location of the ERM.

CONCLUSIONS: We proposed an ERM detection model by synthesizing realistic CFP images with the pathological features of ERM through generative deep learning. We believe that our deep learning framework will help achieve a more accurate detection of ERM in a limited data setting.

PMID:38273286 | DOI:10.1186/s12911-024-02431-4

Phys Med Biol. 2024 Jan 25. doi: 10.1088/1361-6560/ad229f. Online ahead of print.

ABSTRACT

Dual spectral computed tomography (DSCT) is a very challenging problem in the field of imaging. Due to the nonlinearity of its mathematical model, the images reconstructed by the conventional CT usually suffer from the beam hardening artifacts. Additionally, several existing DSCT methods rely heavily on the information of the spectra, which is often not readily available in applications. To address this problem, in this study, we propose a Model-based Direct Inversion Network (MDIN) for DSCT, which can directly predict the basis material images from the collected polychromatic projections. The all operations are performed in the network, requiring
neither the conventional algorithms nor the information of the spectra. It can be viewed as an approximation to the inverse procedure of DSCT imaging model. The MDIN is composed of projection pre-decomposition module (PD-module), domain transformation layer (DT-layer), and image post-decomposition module (ID-module). The PD-module first performs the pre-decomposition on the polychromatic projections that consists of a series of stacked one-dimensional convolution layers. The DT-layer is designed to obtain the preliminary decomposed results, which has the characteristics of sparsely connected and learnable parameters. And the ID-module uses a deep neural network (DNN) to further decompose the reconstructed results of the DT-layer so as to achieve higher-quality basis material images. Numerical experiments demonstrate that the proposed MDIN has significant advantages in substance decomposition,
artifact reduction and noise suppression compared to other methods in the DSCT reconstruction.

PMID:38271738 | DOI:10.1088/1361-6560/ad229f

Phys Med Biol. 2024 Jan 25. doi: 10.1088/1361-6560/ad22a4. Online ahead of print.

ABSTRACT

OBJECTIVE: This paper presents a novel approach for addressing the intricate task of diagnosing aortic valve regurgitation (AR), a valvular disease characterized by blood leakage due to incompetence of the valve closure. Conventional diagnostic techniques require detailed evaluations of multi-modal clinical data, frequently resulting in labor-intensive and time-consuming procedures that are vulnerable to varying subjective assessment of regurgitation severity.

APPROACH: In our research, we introduce the Multi-view Video Contrastive Network (MVCN), designed to leverage multiple color Doppler imaging inputs for multi-view video processing. We leverage supervised contrastive learning as a strategic approach to tackle class imbalance and enhance the effectiveness of our feature representation learning. Specifically, we introduce a contrastive learning framework to enhance representation learning within the embedding space through inter-patient and intra-patient contrastive loss terms.

MAIN RESULTS: We conducted extensive experiments using an in-house dataset comprising 250 echocardiography video series. Our results exhibit a substantial improvement in diagnostic accuracy for AR compared to state-of-the-art methods in terms of accuracy by 9.60%, precision by 8.67%, recall by 9.01%, and F1-score by 8.92%. These results emphasize the capacity of our approach to provide a more precise and efficient method for evaluating the severity of AR.

SIGNIFICANCE: The proposed model could quickly and accurately make decisions about the severity of AR, potentially serving as a useful prescreening tool.

PMID:38271727 | DOI:10.1088/1361-6560/ad22a4

Phys Med Biol. 2024 Jan 25. doi: 10.1088/1361-6560/ad22a1. Online ahead of print.

ABSTRACT

OBJECT: The existing diagnostic paradigm for diabetic retinopathy (DR) greatly relies on subjective assessments by medical practitioners utilizing optical imaging, introducing susceptibility to individual interpretation. This work presents a novel system for the early detection and grading of DR, providing an automated alternative to the manual examination.

APPROACH: First, we use advanced image preprocessing techniques, specifically contrast-limited adaptive histogram equalization and Gaussian filtering, with the goal of enhancing image quality and module learning capabilities. Second, a deep learning-based automatic detection system is developed. The system consists of a feature segmentation module, a deep learning feature extraction module, and an ensemble classification module. The feature segmentation module accomplishes vascular segmentation, the deep learning feature extraction module realizes the global feature and local feature extraction of retinopathy images, and the ensemble module performs the diagnosis and classification of DR for the extracted features. Lastly, nine performance evaluation metrics are applied to assess the quality of the model's performance.

MAIN RESULTS: Extensive experiments are conducted on four retinal image databases (APTOS 2019, Messidor, DDR, and EyePACS). The proposed method demonstrates promising performance in the binary and multi-classification tasks for DR, evaluated through nine indicators, including AUC and quadratic weighted Kappa score. The system shows the best performance in the comparison of three segmentation methods, two convolutional neural network architecture models, four Swin Transformer structures, and the latest literature methods.

SIGNIFICANCE: In contrast to existing methods, our system demonstrates superior performance across multiple indicators, enabling accurate screening of DR and providing valuable support to clinicians in the diagnostic process. Our automated approach minimizes the reliance on subjective assessments, contributing to more consistent and reliable DR evaluations.

PMID:38271723 | DOI:10.1088/1361-6560/ad22a1

Science. 2024 Jan 26;383(6681):406-412. doi: 10.1126/science.adi3794. Epub 2024 Jan 25.

ABSTRACT

We assess which waters the Clean Water Act protects and how Supreme Court and White House rules change this regulation. We train a deep learning model using aerial imagery and geophysical data to predict 150,000 jurisdictional determinations from the Army Corps of Engineers, each deciding regulation for one water resource. Under a 2006 Supreme Court ruling, the Clean Water Act protects two-thirds of US streams and more than half of wetlands; under a 2020 White House rule, it protects less than half of streams and a fourth of wetlands, implying deregulation of 690,000 stream miles, 35 million wetland acres, and 30% of waters around drinking-water sources. Our framework can support permitting, policy design, and use of machine learning in regulatory implementation problems.

PMID:38271507 | DOI:10.1126/science.adi3794

Brief Bioinform. 2024 Jan 22;25(2):bbad528. doi: 10.1093/bib/bbad528.

ABSTRACT

Accurate approaches for quantifying muscle fibers are essential in biomedical research and meat production. In this study, we address the limitations of existing approaches for hematoxylin and eosin-stained muscle fibers by manually and semiautomatically labeling over 660 000 muscle fibers to create a large dataset. Subsequently, an automated image segmentation and quantification tool named MyoV is designed using mask regions with convolutional neural networks and a residual network and feature pyramid network as the backbone network. This design enables the tool to allow muscle fiber processing with different sizes and ages. MyoV, which achieves impressive detection rates of 0.93-0.96 and precision levels of 0.91-0.97, exhibits a superior performance in quantification, surpassing both manual methods and commonly employed algorithms and software, particularly for whole slide images (WSIs). Moreover, MyoV is proven as a powerful and suitable tool for various species with different muscle development, including mice, which are a crucial model for muscle disease diagnosis, and agricultural animals, which are a significant meat source for humans. Finally, we integrate this tool into visualization software with functions, such as segmentation, area determination and automatic labeling, allowing seamless processing for over 400 000 muscle fibers within a WSI, eliminating the model adjustment and providing researchers with an easy-to-use visual interface to browse functional options and realize muscle fiber quantification from WSIs.

PMID:38271484 | DOI:10.1093/bib/bbad528

PLoS One. 2024 Jan 25;19(1):e0292277. doi: 10.1371/journal.pone.0292277. eCollection 2024.

ABSTRACT

Colorectal cancer (CRC) is a major global health concern, with microsatellite instability-high (MSI-H) being a defining characteristic of hereditary nonpolyposis colorectal cancer syndrome and affecting 15% of sporadic CRCs. Tumors with MSI-H have unique features and better prognosis compared to MSI-L and microsatellite stable (MSS) tumors. This study proposed establishing a MSI prediction model using more available and low-cost colonoscopy images instead of histopathology. The experiment utilized a database of 427 MSI-H and 1590 MSS colonoscopy images and vision Transformer (ViT) with different feature training approaches to establish the MSI prediction model. The accuracy of combining pre-trained ViT features was 84% with an area under the receiver operating characteristic curve of 0.86, which was better than that of DenseNet201 (80%, 0.80) in the experiment with support vector machine. The content-based image retrieval (CBIR) approach showed that ViT features can obtain a mean average precision of 0.81 compared to 0.79 of DenseNet201. ViT reduced the issues that occur in convolutional neural networks, including limited receptive field and gradient disappearance, and may be better at interpreting diagnostic information around tumors and surrounding tissues. By using CBIR, the presentation of similar images with the same MSI status would provide more convincing deep learning suggestions for clinical use.

PMID:38271352 | DOI:10.1371/journal.pone.0292277

IEEE Trans Neural Syst Rehabil Eng. 2024 Jan 25;PP. doi: 10.1109/TNSRE.2024.3358497. Online ahead of print.

ABSTRACT

Rehabilitation training is essential for a successful recovery of upper extremity function after stroke. Training programs are typically conducted in hospitals or rehabilitation centers, supervised by specialized medical professionals. However, frequent visits to hospitals can be burdensome for stroke patients with limited mobility. We consider a self-administered rehabilitation system based on a mobile application in which patients can periodically upload videos of themselves performing reach-to-grasp tasks to receive recommendations for self-managed exercises or progress reports. Sensing equipment aside from cameras is typically unavailable in the home environment. A key contribution of our work is to propose a deep learning-based assessment model trained only with video data. As all patients carry out identical tasks, a fine-grained assessment of task execution is required. Our model addresses this difficulty by learning RGB and optical flow data in a complementary manner. The correlation between the RGB and optical flow data is captured by a novel module for modality fusion using cross-attention with Transformers. Experiments showed that our model achieved higher accuracy in movement assessment than existing methods for action recognition. Based on the assessment model, we developed a patient-centered, solution-based mobile application for upper extremity exercises for hemiplegia, which can recommend 57 exercises with three levels of difficulty. A prototype of our application was evaluated by potential end-users and achieved a good quality score on the Mobile Application Rating Scale (MARS).

PMID:38271165 | DOI:10.1109/TNSRE.2024.3358497

Insights Imaging. 2024 Jan 25;15(1):26. doi: 10.1186/s13244-023-01601-8.

ABSTRACT

OBJECTIVES: To use convolutional neural networks (CNNs) for the differentiation between benign and malignant renal tumors using contrast-enhanced CT images of a multi-institutional, multi-vendor, and multicenter CT dataset.

METHODS: A total of 264 histologically confirmed renal tumors were included, from US and Swedish centers. Images were augmented and divided randomly 70%:30% for algorithm training and testing. Three CNNs (InceptionV3, Inception-ResNetV2, VGG-16) were pretrained with transfer learning and fine-tuned with our dataset to distinguish between malignant and benign tumors. The ensemble consensus decision of the three networks was also recorded. Performance of each network was assessed with receiver operating characteristics (ROC) curves and their area under the curve (AUC-ROC). Saliency maps were created to demonstrate the attention of the highest performing CNN.

RESULTS: Inception-ResNetV2 achieved the highest AUC of 0.918 (95% CI 0.873-0.963), whereas VGG-16 achieved an AUC of 0.813 (95% CI 0.752-0.874). InceptionV3 and ensemble achieved the same performance with an AUC of 0.894 (95% CI 0.844-0.943). Saliency maps indicated that Inception-ResNetV2 decisions are based on the characteristics of the tumor while in most tumors considering the characteristics of the interface between the tumor and the surrounding renal parenchyma.

CONCLUSION: Deep learning based on a diverse multicenter international dataset can enable accurate differentiation between benign and malignant renal tumors.

CRITICAL RELEVANCE STATEMENT: Convolutional neural networks trained on a diverse CT dataset can accurately differentiate between benign and malignant renal tumors.

KEY POINTS: • Differentiation between benign and malignant tumors based on CT is extremely challenging. • Inception-ResNetV2 trained on a diverse dataset achieved excellent differentiation between tumor types. • Deep learning can be used to distinguish between benign and malignant renal tumors.

PMID:38270726 | DOI:10.1186/s13244-023-01601-8

Insights Imaging. 2024 Jan 25;15(1):21. doi: 10.1186/s13244-023-01569-5.

ABSTRACT

OBJECTIVE: To establish a model for predicting lymph node metastasis in bladder cancer (BCa) patients.

METHODS: We retroactively enrolled 239 patients who underwent three-phase CT and resection for BCa in two centers (training set, n = 185; external test set, n = 54). We reviewed the clinical characteristics and CT features to identify significant predictors to construct a clinical model. We extracted the hand-crafted radiomics features and deep learning features of the lesions. We used the Minimum Redundancy Maximum Relevance algorithm and the least absolute shrinkage and selection operator logistic regression algorithm to screen features. We used nine classifiers to establish the radiomics machine learning signatures. To compensate for the uneven distribution of the data, we used the synthetic minority over-sampling technique to retrain each machine-learning classifier. We constructed the combined model using the top-performing radiomics signature and clinical model, and finally presented as a nomogram. We evaluated the combined model's performance using the area under the receiver operating characteristic, accuracy, calibration curves, and decision curve analysis. We used the Kaplan-Meier survival curve to analyze the prognosis of BCa patients.

RESULTS: The combined model incorporating radiomics signature and clinical model achieved an area under the receiver operating characteristic of 0.834 (95% CI: 0.659-1.000) for the external test set. The calibration curves and decision curve analysis demonstrated exceptional calibration and promising clinical use. The combined model showed good risk stratification performance for progression-free survival.

CONCLUSION: The proposed CT-based combined model is effective and reliable for predicting lymph node status of BCa patients preoperatively.

CRITICAL RELEVANCE STATEMENT: Bladder cancer is a type of urogenital cancer that has a high morbidity and mortality rate. Lymph node metastasis is an independent risk factor for death in bladder cancer patients. This study aimed to investigate the performance of a deep learning radiomics model for preoperatively predicting lymph node metastasis in bladder cancer patients.

KEY POINTS: • Conventional imaging is not sufficiently accurate to determine lymph node status. • Deep learning radiomics model accurately predicted bladder cancer lymph node metastasis. • The proposed method showed satisfactory patient risk stratification for progression-free survival.

PMID:38270647 | DOI:10.1186/s13244-023-01569-5

J Med Chem. 2024 Jan 25. doi: 10.1021/acs.jmedchem.3c01611. Online ahead of print.

ABSTRACT

Cyclic peptides are gaining attention for their strong binding affinity, low toxicity, and ability to target "undruggable" proteins; however, their therapeutic potential against intracellular targets is constrained by their limited membrane permeability, and researchers need much time and money to test this property in the laboratory. Herein, we propose an innovative multimodal model called Multi_CycGT, which combines a graph convolutional network (GCN) and a transformer to extract one- and two-dimensional features for predicting cyclic peptide permeability. The extensive benchmarking experiments show that our Multi_CycGT model can attain state-of-the-art performance, with an average accuracy of 0.8206 and an area under the curve of 0.8650, and demonstrates satisfactory generalization ability on several external data sets. To the best of our knowledge, it is the first deep learning-based attempt to predict the membrane permeability of cyclic peptides, which is beneficial in accelerating the design of cyclic peptide active drugs in medicinal chemistry and chemical biology applications.

PMID:38270541 | DOI:10.1021/acs.jmedchem.3c01611

Clin Transl Gastroenterol. 2024 Jan 25. doi: 10.14309/ctg.0000000000000681. Online ahead of print.

ABSTRACT

INTRODUCTION: High-resolution anoscopy (HRA) is the gold standard for detecting anal squamous cell cancer (ASCC) precursors. Preliminary studies on the application of artificial intelligence (AI) models to this modality have revealed promising results. However, the impact of staining techniques and anal manipulation on the effectiveness of these algorithms has not been evaluated. We aimed to develop a deep learning system for automatic differentiation of high (HSIL) versus low-grade (LSIL) squamous intraepithelial lesions in HRA images in different subsets of patients (non-stained, acetic acid, lugol, and after manipulation).

METHODS: A convolutional neural network (CNN) was developed to detect and differentiate high and low-grade anal squamous intraepithelial lesions based on 27,770 images from 103 HRA exams performed in 88 patients. Subanalyses were performed to evaluate the algorithm's performance in subsets of images without staining, acetic acid, lugol, and after manipulation of the anal canal. The sensitivity, specificity, accuracy, positive and negative predictive values, and area under the curve (AUC) were calculated.

RESULTS: The CNN achieved an overall accuracy of 98.3%. The algorithm had a sensitivity and specificity of 97.4% and 99.2%, respectively. The accuracy of the algorithm for differentiating HSIL vs LSIL varied between 91.5% (post-manipulation) and 100% (lugol) for the categories at subanalysis. The AUC ranged between 0.95 and 1.00.

DISCUSSION: The introduction of AI to HRA may provide an accurate detection and differentiation of ASCC precursors. Our algorithm showed excellent performance at different staining settings. This is extremely important as real-time AI models during HRA exams can help guide local treatment or detect relapsing disease.

PMID:38270249 | DOI:10.14309/ctg.0000000000000681

J Chem Inf Model. 2024 Jan 25. doi: 10.1021/acs.jcim.3c01747. Online ahead of print.

ABSTRACT

Thermally activated delayed fluorescence (TADF) material has attracted great attention as a promising metal-free organic light-emitting diode material with a high theoretical efficiency. To accelerate the discovery of novel TADF materials, computer-aided material design strategies have been developed. However, they have clear limitations due to the accessibility of only a few computationally tractable properties. Here, we propose TADF-likeness, a quantitative score to evaluate the TADF potential of molecules based on a data-driven concept of chemical similarity to existing TADF molecules. We used a deep autoencoder to characterize the common features of existing TADF molecules with common chemical descriptors. The score was highly correlated with the four essential electronic properties of TADF molecules and had a high success rate in large-scale virtual screening of millions of molecules to identify promising candidates at almost no cost, validating its feasibility for accelerating TADF discovery. The concept of TADF-likeness can be extended to other fields of materials discovery.

PMID:38270063 | DOI:10.1021/acs.jcim.3c01747

Stud Health Technol Inform. 2024 Jan 25;310:1354-1355. doi: 10.3233/SHTI231191.

NO ABSTRACT

PMID:38270040 | DOI:10.3233/SHTI231191

J Appl Clin Med Phys. 2024 Jan 25:e14266. doi: 10.1002/acm2.14266. Online ahead of print.

ABSTRACT

PURPOSE: Non-Contrast Enhanced CT (NCECT) is normally required for proton dose calculation while Contrast Enhanced CT (CECT) is often scanned for tumor and organ delineation. Possible tissue motion between these two CTs raises dosimetry uncertainties, especially for moving tumors in the thorax and abdomen. Here we report a deep-learning approach to generate NCECT directly from CECT. This method could be useful to avoid the NCECT scan, reduce CT simulation time and imaging dose, and decrease the uncertainties caused by tissue motion between otherwise two different CT scans.

METHODS: A deep network was developed to convert CECT to NCECT. The network receives a 3D image from CECT images as input and generates a corresponding contrast-removed NCECT image patch. Abdominal CECT and NCECT image pairs of 20 patients were deformably registered and 8000 image patch pairs extracted from the registered image pairs were utilized to train and test the model. CTs of clinical proton patients and their treatment plans were employed to evaluate the dosimetric impact of using the generated NCECT for proton dose calculation.

RESULTS: Our approach achieved a Cosine Similarity score of 0.988 and an MSE value of 0.002. A quantitative comparison of clinical proton dose plans computed on the CECT and the generated NCECT for five proton patients revealed significant dose differences at the distal of beam paths. V100% of PTV and GTV changed by 3.5% and 5.5%, respectively. The mean HU difference for all five patients between the generated and the scanned NCECTs was ∼4.72, whereas the difference between CECT and the scanned NCECT was ∼64.52, indicating a ∼93% reduction in mean HU difference.

CONCLUSIONS: A deep learning approach was developed to generate NCECTs from CECTs. This approach could be useful for the proton dose calculation to reduce uncertainties caused by tissue motion between CECT and NCECT.

PMID:38269961 | DOI:10.1002/acm2.14266

Stud Health Technol Inform. 2024 Jan 25;310:936-940. doi: 10.3233/SHTI231102.

ABSTRACT

Microvascular invasion of HCC is an important factor affecting postoperative recurrence and prognosis of patients. Preoperative diagnosis of MVI is greatly significant to improve the prognosis of HCC. Currently, the diagnosis of MVI is mainly based on the histopathological examination after surgery, which is difficult to meet the requirement of preoperative diagnosis. Also, the sensitivity, specificity and accuracy of MVI diagnosis based on a single imaging feature are low. In this paper, a robust, high-precision cross-modality unified framework for clinical diagnosis is proposed for the prediction of microvascular invasion of hepatocellular carcinoma. It can effectively extract, fuse and locate multi-phase MR Images and clinical data, enrich the semantic context, and comprehensively improve the prediction indicators in different hospitals. The state-of-the-art performance of the approach was validated on a dataset of HCC patients with confirmed pathological types. Moreover, CMIR provides a possible solution for related multimodality tasks in the medical field.

PMID:38269946 | DOI:10.3233/SHTI231102

Stud Health Technol Inform. 2024 Jan 25;310:926-930. doi: 10.3233/SHTI231100.

ABSTRACT

Survival prediction is crucial for treatment decision making in hepatocellular carcinoma (HCC). We aimed to build a fully automated artificial intelligence system (FAIS) that mines whole-liver information to predict overall survival of HCC. We included 215 patients with preoperative contrast-enhance CT imaging and received curative resection from a hospital in China. The cohort was randomly split into developing and testing subcohorts. The FAIS was constructed with convolutional layers and full-connected layers. Cox regression loss was used for training. Models based on clinical and/or tumor-based radiomics features were built for comparison. The FAIS achieved C-indices of 0.81 and 0.72 for the developing and testing sets, outperforming all the other three models. In conclusion, our study suggest that more important information could be mined from whole liver instead of only the tumor. Our whole-liver based FAIS provides a non-invasive and efficient overall survival prediction tool for HCC before the surgery.

PMID:38269944 | DOI:10.3233/SHTI231100