An Automated Deep Learning-Based Framework for Uptake Segmentation and Classification on PSMA PET/CT Imaging of Patients with Prostate Cancer.

Uptake segmentation and classification on PSMA PET/CT are important for automating whole-body tumor burden determinations. We developed and evaluated an automated deep learning (DL)-based framework that segments and classifies uptake on PSMA PET/CT. We identified 193 [18F] DCFPyL PET/CT scans of patients with biochemically recurrent prostate cancer from two institutions, including 137 [18F] DCFPyL PET/CT scans for training and internally testing, and 56 scans from another institution for external testing. Two radiologists segmented and labelled foci as suspicious or non-suspicious for malignancy. A DL-based segmentation was developed with two independent CNNs. An anatomical prior guidance was applied to make the DL framework focus on PSMA-avid lesions. Segmentation performance was evaluated by Dice, IoU, precision, and recall. Classification model was constructed with multi-modal decision fusion framework evaluated by accuracy, AUC, F1 score, precision, and recall. Automatic segmentation of suspicious lesions was improved under prior guidance, with mean Dice, IoU, precision, and recall of 0.700, 0.566, 0.809, and 0.660 on the internal test set and 0.680, 0.548, 0.749, and 0.740 on the external test set. Our multi-modal decision fusion framework outperformed single-modal and multi-modal CNNs with accuracy, AUC, F1 score, precision, and recall of 0.764, 0.863, 0.844, 0.841, and 0.847 in distinguishing suspicious and non-suspicious foci on the internal test set and 0.796, 0.851, 0.865, 0.814, and 0.923 on the external test set. DL-based lesion segmentation on PSMA PET is facilitated through our anatomical prior guidance strategy. Our classification framework differentiates suspicious foci from those not suspicious for cancer with good accuracy.

Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction.

Deep learning models have demonstrated great potential in medical imaging but are limited by the expensive, large volume of annotations required. To address this, we compared different active learning strategies by training models on subsets of the most informative images using real-world clinical datasets for brain tumor segmentation and proposing a framework that minimizes the data needed while maintaining performance. Then, 638 multi-institutional brain tumor magnetic resonance imaging scans were used to train three-dimensional U-net models and compare active learning strategies. Uncertainty estimation techniques including Bayesian estimation with dropout, bootstrapping, and margins sampling were compared to random query. Strategies to avoid annotating similar images were also considered. We determined the minimum data necessary to achieve performance equivalent to the model trained on the full dataset (α = 0.05). Bayesian approximation with dropout at training and testing showed results equivalent to that of the full data model (target) with around 30% of the training data needed by random query to achieve target performance (p = 0.018). Annotation redundancy restriction techniques can reduce the training data needed by random query to achieve target performance by 20%. We investigated various active learning strategies to minimize the annotation burden for three-dimensional brain tumor segmentation. Dropout uncertainty estimation achieved target performance with the least annotated data.

Swin-Net: A Swin-Transformer-Based Network Combing with Multi-Scale Features for Segmentation of Breast Tumor Ultrasound Images.

Breast cancer is one of the most common cancers in the world, especially among women. Breast tumor segmentation is a key step in the identification and localization of the breast tumor region, which has important clinical significance. Inspired by the swin-transformer model with powerful global modeling ability, we propose a semantic segmentation framework named Swin-Net for breast ultrasound images, which combines Transformer and Convolutional Neural Networks (CNNs) to effectively improve the accuracy of breast ultrasound segmentation. Firstly, our model utilizes a swin-transformer encoder with stronger learning ability, which can extract features of images more precisely. In addition, two new modules are introduced in our method, including the feature refinement and enhancement module (RLM) and the hierarchical multi-scale feature fusion module (HFM), given that the influence of ultrasonic image acquisition methods and the characteristics of tumor lesions is difficult to capture. Among them, the RLM module is used to further refine and enhance the feature map learned by the transformer encoder. The HFM module is used to process multi-scale high-level semantic features and low-level details, so as to achieve effective cross-layer feature fusion, suppress noise, and improve model segmentation performance. Experimental results show that Swin-Net performs significantly better than the most advanced methods on the two public benchmark datasets. In particular, it achieves an absolute improvement of 1.4-1.8% on Dice. Additionally, we provide a new dataset of breast ultrasound images on which we test the effect of our model, further demonstrating the validity of our method. In summary, the proposed Swin-Net framework makes significant advancements in breast ultrasound image segmentation, providing valuable exploration for research and applications in this domain.

Catalytic hydrolysis of carbonyl sulfide in blast furnace gas over Sm-Ce-Ox@ZrO2 catalyst.

Carbonyl sulfur (COS) is a prominent organic sulfur pollutant commonly found in the by-product gas generated by the steel industry. A series of Sm-doped CeOx@ZrO2 catalysts were prepared for the hydrolysis catalytic removal of COS. The results showed that the addition of Sm resulted in the most significant enhancement of hydrolysis catalytic activity. The 3% Sm2O3-Ce-Ox@ZrO2 catalyst exhibited the highest activity, achieving a hydrolysis catalytic efficiency of 100% and H2S selectivity of 100% within the temperature range of 90-180 °C. The inclusion of Sm had the effect of reducing the acidity of the catalyst while increasing weak basic sites, which facilitated the adsorption and activation of COS molecules at low temperatures. Appropriate doping of Sm proved beneficial in converting active surface chemisorbed oxygen into lattice oxygen, thereby decreasing the oxidation of intermediate products and maintaining the stability of the hydrolysis reaction.

Improvement of sleep quality in isolated metastatic patients with spinal cord compression after surgery.

BACKGROUND: This study aimed to assess changes in quality of sleep (QoS) in isolated metastatic patients with spinal cord compression following two different surgical treatments and identify potential contributing factors associated with QoS improvement. METHODS: We reviewed 49 patients with isolated spinal metastasis at our spinal tumor center between December 2017 and May 2021. Total en bloc spondylectomy (TES) and palliative surgery with postoperative stereotactic radiosurgery (PSRS) were performed on 26 and 23 patients, respectively. We employed univariate and multivariate analyses to identify the potential prognostic factors affecting QoS. RESULTS: The total Pittsburgh Sleep Quality Index (PSQI) score improved significantly 6 months after surgery. Univariate analysis indicated that age, pain worsening at night, decrease in visual analog scale (VAS), increase in Eastern Cooperative Oncology Group performance score (ECOG-PS), artificial implant in focus, and decrease in epidural spinal cord compression (ESCC) scale values were potential contributing factors for QoS. Multivariate analysis indicated that the ESCC scale score decreased as an independent prognostic factor. CONCLUSIONS: Patients with spinal cord compression caused by the metastatic disease had significantly improved QoS after TES and PSRS treatment. Moreover, a decrease in ESCC scale value of > 1 was identified as a favorable contributing factor associated with PSQI improvement. In addition, TES and PSRS can prevent recurrence by achieving efficient local tumor control to improve indirect sleep. Accordingly, timely and effective surgical decompression and recurrence control are critical for improving sleep quality.

Automated machine learning for differentiation of hepatocellular carcinoma from intrahepatic cholangiocarcinoma on multiphasic MRI.

With modern management of primary liver cancer shifting towards non-invasive diagnostics, accurate tumor classification on medical imaging is increasingly critical for disease surveillance and appropriate targeting of therapy. Recent advancements in machine learning raise the possibility of automated tools that can accelerate workflow, enhance performance, and increase the accessibility of artificial intelligence to clinical researchers. We explore the use of an automated Tree-Based Optimization Tool that leverages a genetic programming algorithm for differentiation of the two common primary liver cancers on multiphasic MRI. Manual and automated analyses were performed to select an optimal machine learning model, with an accuracy of 73-75% (95% CI 0.59-0.85), sensitivity of 70-75% (95% CI 0.48-0.89), and specificity of 71-79% (95% CI 0.52-0.90) on manual optimization, and an accuracy of 73-75% (95% CI 0.59-0.85), sensitivity of 65-75% (95% CI 0.43-0.89) and specificity of 75-79% (95% CI 0.56-0.90) for automated machine learning. We found that automated machine learning performance was similar to that of manual optimization, and it could classify hepatocellular carcinoma and intrahepatic cholangiocarcinoma with an sensitivity and specificity comparable to that of radiologists. However, automated machine learning performance was poor on a subset of scans that met LI-RADS criteria for LR-M. Exploration of additional feature selection and classifier methods with automated machine learning to improve performance on LR-M cases as well as prospective validation in the clinical setting are needed prior to implementation.

Deep learning-based automatic tumor burden assessment of pediatric high-grade gliomas, medulloblastomas, and other leptomeningeal seeding tumors.

BACKGROUND: Longitudinal measurement of tumor burden with magnetic resonance imaging (MRI) is an essential component of response assessment in pediatric brain tumors. We developed a fully automated pipeline for the segmentation of tumors in pediatric high-grade gliomas, medulloblastomas, and leptomeningeal seeding tumors. We further developed an algorithm for automatic 2D and volumetric size measurement of tumors. METHODS: The preoperative and postoperative cohorts were randomly split into training and testing sets in a 4:1 ratio. A 3D U-Net neural network was trained to automatically segment the tumor on T1 contrast-enhanced and T2/FLAIR images. The product of the maximum bidimensional diameters according to the RAPNO (Response Assessment in Pediatric Neuro-Oncology) criteria (AutoRAPNO) was determined. Performance was compared to that of 2 expert human raters who performed assessments independently. Volumetric measurements of predicted and expert segmentations were computationally derived and compared. RESULTS: A total of 794 preoperative MRIs from 794 patients and 1003 postoperative MRIs from 122 patients were included. There was excellent agreement of volumes between preoperative and postoperative predicted and manual segmentations, with intraclass correlation coefficients (ICCs) of 0.912 and 0.960 for the 2 preoperative and 0.947 and 0.896 for the 2 postoperative models. There was high agreement between AutoRAPNO scores on predicted segmentations and manually calculated scores based on manual segmentations (Rater 2 ICC = 0.909; Rater 3 ICC = 0.851). Lastly, the performance of AutoRAPNO was superior in repeatability to that of human raters for MRIs with multiple lesions. CONCLUSIONS: Our automated deep learning pipeline demonstrates potential utility for response assessment in pediatric brain tumors. The tool should be further validated in prospective studies.

Performance of automatic machine learning versus radiologists in the evaluation of endometrium on computed tomography.

PURPOSE: In this study, we developed radiomic models that utilize a combination of imaging features and clinical variables to distinguish endometrial cancer (EC) from normal endometrium on routine computed tomography (CT). METHODS: A total of 926 patients consisting of 416 endometrial cancer (EC) and 510 normal endometrium were included. The CT images of these patients were segmented manually, and divided into training, validation, testing and external testing sets. Non-texture and texture features of these images with endometrium or uterus as region of interest were extracted. The clinical feature "age" was also included in the feature set. Feature selection and machine learning classifier were applied to normalized feature set. This manual optimized combination was then compared with the best pipeline exported by Tree-Based Pipeline Optimization Tool (TPOT) on testing and external testing set. The performances of these machine learning pipelines were compared to that of radiologists. RESULTS: The manual expert optimized pipeline using the "reliefF" feature selection method and "Bagging" classifier on the external testing set achieved a test ROC AUC of 0.73, accuracy of 0.73 (95% CI 0.62-0.82), sensitivity of 0.64 (95% CI 0.45-0.79), and specificity of 0.78 (95% CI 0.65-0.87), while TPOT achieved a test ROC AUC of 0.79, accuracy of 0.80 (95% CI 0.70-0.87), sensitivity of 0.61 (95% CI 0.43-0.77), and specificity of 0.90 (95% CI 0.78-0.96). When compared to average radiologist performance, the TPOT achieved higher test accuracy (0.80 vs. 0.49, p < 0.001) and specificity (0.90 vs. 0.51, p < 0.001), with comparable sensitivity (0.61 vs. 0.46, p = 0.130). CONCLUSION: Our results demonstrate that automatic machine learning can distinguish EC from normal endometrium on routine CT imaging with higher accuracy and specificity than radiologists.

Development and Validation of a Scoring System for Differential Diagnosis of Tuberculosis and Metastatic Tumor in the Spine.

PURPOSE: Spinal tuberculosis (TB) and metastatic tumor (MT) are common diseases with similar manifestations. Although pathological evaluation is the gold standard to confirm diagnosis, performing biopsies in all patients is not feasible. This study is aimed to create a scoring system to facilitate the differential diagnosis of spinal TB and MT before invasive procedures. METHODS: Altogether, 447 patients with spinal TB (n=198) and MT (n=249) were retrospectively analyzed. Patients were randomly assigned at 2:1 ratio to a training cohort and a validation cohort. Clinical, laboratory, and radiological diagnostic factors were identified by χ2 and multiple logistic regression analyses. The scoring system was then established based on the identified independent diagnostic factors scored by regression coefficient ß value, with the cut-off value being determined by ROC curve. The sensitivity and specificity of the system was calculated by comparing the predicted diagnosis with their actual pathological diagnosis. RESULTS: This scoring system was composed of 5 items: pain worsens at night (0 or 2 points), CRP value (0 or 3 points), tumor marker values (0 or 2 points), skip lesions (0 or 3 points), and intervertebral space destruction (0 or 3 points). Patients scoring higher than 7.5 could be diagnosed as spinal TB, otherwise, MT. According to the internal validation, the sensitivity and specificity of the system were 87.9% and 91.6%, respectively. CONCLUSION: This study established and validated a scoring system which could be used to differentiate spinal TB from MT, thus helping clinicians in quick and accurate differential diagnosis.

Preoperative prediction of the stage, size, grade, and necrosis score in clear cell renal cell carcinoma using MRI-based radiomics.

PURPOSE: Clear cell renal cell carcinoma (ccRCC) is the most common subtype of renal cell carcinoma. Currently, there is a lack of noninvasive methods to stratify ccRCC prognosis prior to any invasive therapies. The purpose of this study was to preoperatively predict the tumor stage, size, grade, and necrosis (SSIGN) score of ccRCC using MRI-based radiomics. METHODS: A multicenter cohort of 364 histopathologically confirmed ccRCC patients (272 low [< 4] and 92 high [≥ 4] SSIGN score) with preoperative T2-weighted and T1-contrast-enhanced MRI were retrospectively identified and divided into training (254 patients) and testing sets (110 patients). The performance of a manually optimized radiomics model was assessed by measuring accuracy, sensitivity, specificity, area under receiver operating characteristic curve (AUROC), and area under precision-recall curve (AUPRC) on an independent test set, which was not included in model training. Lastly, its performance was compared to that of a machine learning pipeline, Tree-Based Pipeline Optimization Tool (TPOT). RESULTS: The manually optimized radiomics model using Random Forest classification and Analysis of Variance feature selection methods achieved an AUROC of 0.89, AUPRC of 0.81, accuracy of 0.89 (95% CI 0.816-0.937), specificity of 0.95 (95% CI 0.875-0.984), and sensitivity of 0.72 (95% CI 0.537-0.852) on the test set. The TPOT using Extra Trees Classifier achieved an AUROC of 0.94, AUPRC of 0.83, accuracy of 0.89 (95% CI 0.816-0.937), specificity of 0.95 (95% CI 0.875-0.984), and sensitivity of 0.72 (95% CI 0.537-0.852) on the test set. CONCLUSION: Preoperative MR radiomics can accurately predict SSIGN score of ccRCC, suggesting its promise as a prognostic tool that can be used in conjunction with diagnostic markers.

Ossifying Fibroma of the Cervical Spine: First Case Report and Literature Review.

BACKGROUND: Ossifying fibroma (OF) is a benign tumor commonly occurring in the mandible and maxilla. Spinal involvement of OFs is of great rarity. To the best of our knowledge, only 3 cases in the thoracic and lumbar spine have been reported. CASE DESCRIPTION: We present the case of a 22-year-old woman with an OF of the atlas, which, to the best of our knowledge, is the first described OF in the cervical spine. The related data were also reviewed. Only 3 spinal OFs involving the thoracic spine to the sacrum have been reported. CONCLUSIONS: Spinal involvement of OFs seldom occurs. To the best of our knowledge, we have reported the first OF involving the cervical spine. Differentiating OFs from primary spinal tumors is necessary. OFs have locally aggressive behavior and a high risk of recurrence. Complete resection, followed by regular examinations, should be the ideal choice for treatment.

A Survey on Intelligent Screening for Diabetic Retinopathy.

Diabetic retinopathy (DR) is one of the leading causes of vision loss and can be effectively avoided by screening, early diagnosis and treatment. In order to increase the universality and efficiency of DR screening, many efforts have been invested in developing intelligent screening, and there have been great advances. In this paper, we survey DR screening from four perspectives: 1) public color fundus image datasets of DR; 2) DR classification and related lesion-extraction approaches; 3) existing computer-aided systems for DR screening; and 4) existing issues, challenges, and research trends. Our goal is to provide insights for future research directions on DR intelligent screening.

L-cysteine-assisted synthesis of hierarchical NiS2 hollow spheres supported carbon nitride as photocatalysts with enhanced lifetime.

Novel hierarchical NiS2 hollow spheres modified by graphite-like carbon nitride were prepared using a facile L-cysteine-assisted solvothermal route. The NiS2/g-C3N4 composites exhibited excellent photocatalytic efficiency in rhodamine B, methyl orange and ciprofloxacin degradation as compared to single g-C3N4 and NiS2, which could be due to the synergistic effects of the unique hollow sphere-like structure, strong visible-light absorption and increased separation rate of the photoinduced electron-hole pairs at the intimate interface of heterojunctions. A suitable combination of g-C3N4 with NiS2 showed the best photocatalytic performance. In addition, an electron spin resonance and trapping experiment demonstrated that the photogenerated hydroxyl radicals and superoxide radicals were the two main photoactive species in photocatalysis. A possible photocatalytic mechanism of NiS2/g-C3N4 composites under visible light irradiation is also proposed. The strategy presented here can be extended to a general strategy for constructing 3D/2D heterostructured photocatalysts for broad applications in photocatalysis.

Automatic cytoplasm and nuclei segmentation for color cervical smear image using an efficient gap-search MRF.

Accurate and effective cervical smear image segmentation is required for automated cervical cell analysis systems. Thus, we proposed a novel superpixel-based Markov random field (MRF) segmentation framework to acquire the nucleus, cytoplasm and image background of cell images. We seek to classify color non-overlapping superpixel-patches on one image for image segmentation. This model describes the whole image as an undirected probabilistic graphical model and was developed using an automatic label-map mechanism for determining nuclear, cytoplasmic and background regions. A gap-search algorithm was designed to enhance the model efficiency. Data show that the algorithms of our framework provide better accuracy for both real-world and the public Herlev datasets. Furthermore, the proposed gap-search algorithm of this model is much more faster than pixel-based and superpixel-based algorithms.