Experimental Boson Sampling Enabling Cryptographic One-Way Function.

Boson sampling is a computational problem, which is commonly believed to be a representative paradigm for attaining the milestone of quantum advantage. So far, massive efforts have been made to the experimental large-scale boson sampling for demonstrating this milestone, while further applications of the machines remain a largely unexplored area. Here, we investigate experimentally the efficiency and security of a cryptographic one-way function that relies on coarse-grained boson sampling, in the framework of a photonic boson-sampling machine fabricated by a femtosecond laser direct writing technique. Our findings demonstrate that the implementation of the function requires moderate sample sizes, which can be over 4 orders of magnitude smaller than the ones predicted by the Chernoff bound; whereas for numbers of photons n≥3 and bins d∼poly(m,n), the same output of the function cannot be generated by nonboson samplers. Our Letter is the first experimental study that deals with the potential applications of boson sampling in the field of cryptography and paves the way toward additional studies in this direction.

Robust deep learning-based PET prognostic imaging biomarker for DLBCL patients: a multicenter study.

OBJECTIVE: To develop and independently externally validate robust prognostic imaging biomarkers distilled from PET images using deep learning techniques for precise survival prediction in patients with diffuse large B cell lymphoma (DLBCL). METHODS: A total of 684 DLBCL patients from three independent medical centers were included in this retrospective study. Deep learning scores (DLS) were generated from PET images using deep convolutional neural network architecture known as VGG19 and DenseNet121. These DLSs were utilized to predict progression-free survival (PFS) and overall survival (OS). Furthermore, multiparametric models were designed based on results from the Cox proportional hazards model and assessed through calibration curves, concordance index (C-index), and decision curve analysis (DCA) in the training and validation cohorts. RESULTS: The DLSPFS and DLSOS exhibited significant associations with PFS and OS, respectively (P<0.05) in the training and validation cohorts. The multiparametric models that incorporated DLSs demonstrated superior efficacy in predicting PFS (C-index: 0.866) and OS (C-index: 0.835) compared to competing models in training cohorts. In external validation cohorts, the C-indices for PFS and OS were 0.760 and. 0.770 and 0.748 and 0.766, respectively, indicating the reliable validity of the multiparametric models. The calibration curves displayed good consistency, and the decision curve analysis (DCA) confirmed that the multiparametric models offered more net clinical benefits. CONCLUSIONS: The DLSs were identified as robust prognostic imaging biomarkers for survival in DLBCL patients. Moreover, the multiparametric models developed in this study exhibited promising potential in accurately stratifying patients based on their survival risk.

Prognostic Value of Left Atrial Reservoir Strain in Left Ventricular Myocardial Noncompaction: A 3.0 T Cardiac Magnetic Resonance Feature Tracking Study.

BACKGROUND: The relationship of left atrial (LA) strain to high-risk heart failure (HF) events in patients with left ventricular myocardial noncompaction (LVNC) remains to be thoroughly investigated. PURPOSE: To evaluate the LA performance in patients with LVNC, and to investigate the prognostic value of LA phasic strain on high-risk HF events, and its influencing factors. STUDY TYPE: Retrospective. POPULATION: A total of 95 LVNC patients (74 with LA enlargement [LAE] and 21 without LAE) and 50 healthy controls. FIELD STRENGTH/SEQUENCE: A 3.0 T, balanced steady-state free-precession cine imaging. ASSESSMENT: LA longitudinal strains were measured by cardiac MRI feature tracking technique. LA volume index (LAVI) and LA ejection fraction (LAEF) were calculated. Their intraobserver and interobserver reproducibility were evaluated. The primary outcome was high-risk HF events, a composite of first HF hospitalization, hospitalization for worsening HF and death from HF. STATISTICAL TESTS: Student's t/Mann-Whitney U, one-way analysis of variance/Kruskal-Wallis, Chi-squared, receiver operating characteristic, Kaplan-Meier, log-rank, Cox regression, Pearson and Spearman correlation and linear regression analyses were performed. The significance threshold was set at P < 0 .05. RESULTS: LAEF and LA longitudinal strains decreased in LVNC patients irrespective of the presence of LAE. During a median follow-up of 32.17 months, high-risk HF occurred in 13 (13.68%) patients. Patients with increased LAVI, decreased LAEF and decreased LA longitudinal strain had significantly higher risks of high-risk HF events. In patients with LVNC, LA reservoir strain (εs) was independently associated with high-risk HF (hazard ratio = 23.208 [95% CI: 2.993-179.967]). LV global longitudinal strain (LV GLS) (ß = -1.783 [95% CI: -2.493 to -1.073]) was significantly and independently associated with εs. Intraobserver and interobserver reproducibility was excellent for LAVI, LAEF, and LA strain. CONCLUSION: In patients with LVNC, εs was an independent predictor for high-risk HF events. LV GLS was an independent determinant of εs in LVNC. EVIDENCE LEVEL: 4 TECHNICAL EFFICACY: Stage 4.

Musculoskeletal Ultrasound Image-Based Radiomics for the Diagnosis of Achilles Tendinopathy in Skiers.

OBJECTIVES: Our study aimed to develop and validate an efficient ultrasound image-based radiomic model for determining the Achilles tendinopathy in skiers. METHODS: A total of 88 feet of skiers clinically diagnosed with unilateral chronic Achilles tendinopathy and 51 healthy feet were included in our study. According to the time order of enrollment, the data were divided into a training set (n = 89) and a test set (n = 50). The regions of interest (ROIs) were segmented manually, and 833 radiomic features were extracted from red, green, blue color channels and grayscale of ROIs using Pyradiomics, respectively. Three feature selection and three machine learning modeling algorithms were implemented respectively, for determining the optimal radiomics pipeline. Finally, the area under the receiver operating characteristic curve (AUC), consistency analysis, and decision analysis were used to evaluate the diagnostic performance. RESULTS: By comparing nine radiomics analysis strategies of three color channels and grayscale, the radiomic model under the green channel obtained the best diagnostic performance, using the Random Forest selection and Support Vector Machine modeling, which was selected as the final machine learning model. All the selected radiomic features were significantly associated with the Achilles tendinopathy (P < .05). The radiomic model had a training AUC of 0.98, a test AUC of 0.99, a sensitivity of 0.90, and a specificity of 1, which could bring sufficient clinical net benefits. CONCLUSIONS: Ultrasound image-based radiomics achieved high diagnostic performance, which could be used as an intelligent auxiliary tool for the diagnosis of Achilles tendinopathy.

Direct Observation of Dynamically Localized Quantum Optical States.

Quantum-correlated biphoton states play an important role in quantum communication and processing, especially considering the recent advances in integrated photonics. However, it remains a challenge to flexibly transport quantum states on a chip, when dealing with large-scale sophisticated photonic designs. The equivalence between certain aspects of quantum optics and solid-state physics makes it possible to utilize a range of powerful approaches in photonics, including topologically protected boundary states, graphene edge states, and dynamic localization. Optical dynamic localization allows efficient protection of classical signals in photonic systems by implementing an analogue of an external alternating electric field. Here, we report on the observation of dynamic localization for quantum-correlated biphotons, including both the generation and the propagation aspects. As a platform, we use sinusoidal waveguide arrays with cubic nonlinearity. We record biphoton coincidence count rates as evidence of robust generation of biphotons and demonstrate the dynamic localization features in both spatial and temporal space by analyzing the quantum correlation of biphotons at the output of the waveguide array. Experimental results demonstrate that various dynamic modulation parameters are effective in protecting quantum states without introducing complex topologies. Our Letter opens new avenues for studying complex physical processes using photonic chips and provides an alternative mechanism of protecting communication channels and nonclassical quantum sources in large-scale integrated quantum optics.

Multimodality MRI-based radiomics approach to predict the posttreatment response of lung cancer brain metastases to gamma knife radiosurgery.

OBJECTIVES: To develop and validate a multimodality MRI-based radiomics approach to predicting the posttreatment response of lung cancer brain metastases (LCBM) to gamma knife radiosurgery (GKRS). METHODS: We retrospectively analyzed 213 lesions from 137 patients with LCBM who received GKRS between January 2017 and November 2020. The data were divided into a primary cohort (102 patients with 173 lesions) and an independent validation cohort (35 patients with 40 lesions) according to the time of treatment. Benefit result was defined using pretreatment and 3-month follow-up MRI images based on the Response Assessment in Neuro-Oncology Brain Metastases criteria. Valuable radiomics features were extracted from pretreatment multimodality MRI images using random forests. Prediction performance among the radiomics features of tumor core (RFTC) and radiomics features of peritumoral edema (RFPE) together was evaluated separately. Then, the random forest radiomics score and nomogram were developed through the primary cohort and evaluated through an independent validation cohort. Prediction performance was evaluated by ROC curve, calibration curve, and decision curve. RESULTS: Gender (p = 0.018), histological subtype (p = 0.009), epidermal growth factor receptor mutation (p = 0.034), and targeted drug treatment (p = 0.021) were significantly associated with posttreatment response. Adding RFPE to RFTC showed improved prediction performance than RFTC alone in primary cohort (AUC = 0.848 versus AUC = 0.750; p < 0.001). Finally, the radiomics nomogram had an AUC of 0.930, a C-index of 0.930 (specificity of 83.1%, sensitivity of 87.3%) in primary cohort, and an AUC of 0.852, a C-index of 0.848 (specificity of 84.2%, sensitivity of 76.2%) in validation cohort. CONCLUSIONS: Multimodality MRI-based radiomics models can predict the posttreatment response of LCBM to GKRS. KEY POINTS: • Among the selected radiomics features, texture features basically contributed the dominant force in prediction tasks (80%), especially gray-level co-occurrence matrix features (40%). • Adding RFPE to RFTC showed improved prediction performance than RFTC alone in primary cohort (AUC = 0.848 versus AUC = 0.750; p < 0.001). • The multimodality MRI-based radiomics nomogram showed high accuracy for distinguishing the posttreatment response of LCBM to GKRS (AUC = 0.930, in primary cohort; AUC = 0.852, in validation cohort).

Diagnostic accuracy and potential covariates of artificial intelligence for diagnosing orthopedic fractures: a systematic literature review and meta-analysis.

OBJECTIVES: To systematically quantify the diagnostic accuracy and identify potential covariates affecting the performance of artificial intelligence (AI) in diagnosing orthopedic fractures. METHODS: PubMed, Embase, Web of Science, and Cochrane Library were systematically searched for studies on AI applications in diagnosing orthopedic fractures from inception to September 29, 2021. Pooled sensitivity and specificity and the area under the receiver operating characteristic curves (AUC) were obtained. This study was registered in the PROSPERO database prior to initiation (CRD 42021254618). RESULTS: Thirty-nine were eligible for quantitative analysis. The overall pooled AUC, sensitivity, and specificity were 0.96 (95% CI 0.94-0.98), 90% (95% CI 87-92%), and 92% (95% CI 90-94%), respectively. In subgroup analyses, multicenter designed studies yielded higher sensitivity (92% vs. 88%) and specificity (94% vs. 91%) than single-center studies. AI demonstrated higher sensitivity with transfer learning (with vs. without: 92% vs. 87%) or data augmentation (with vs. without: 92% vs. 87%), compared to those without. Utilizing plain X-rays as input images for AI achieved results comparable to CT (AUC 0.96 vs. 0.96). Moreover, AI achieved comparable results to humans (AUC 0.97 vs. 0.97) and better results than non-expert human readers (AUC 0.98 vs. 0.96; sensitivity 95% vs. 88%). CONCLUSIONS: AI demonstrated high accuracy in diagnosing orthopedic fractures from medical images. Larger-scale studies with higher design quality are needed to validate our findings. KEY POINTS: • Multicenter study design, application of transfer learning, and data augmentation are closely related to improving the performance of artificial intelligence models in diagnosing orthopedic fractures. • Utilizing plain X-rays as input images for AI to diagnose fractures achieved results comparable to CT (AUC 0.96 vs. 0.96). • AI achieved comparable results to humans (AUC 0.97 vs. 0.97) but was superior to non-expert human readers (AUC 0.98 vs. 0.96, sensitivity 95% vs. 88%) in diagnosing fractures.

A fast fluorescent probe for tracing endoplasmic reticulum-located carboxylesterase in living cells.

Carboxylesterase (CEs), mainly localized in endoplasmic reticulum (ER), are responsible for hydrolyzing compounds containing various ester bonds. They have been closely associated with drug metabolism and cellular homeostasis. Although some CE fluorescent probes have been developed, there are still a lack of probes that could target to the ER. Here, we developed a novel fluorescent probe CR with a specific ER anchor for monitoring CEs. In CR, p-toluenesulfonamide was chosen for precise ER targeting. A simple acetyl moiety was used as the CE response site and fluorescence modulation unit. During the spectral tests, CR displayed a fast response speed (within 10 s) towards CEs. In addition, it showed high sensitivity [limit of detection (LOD) = 5.1 × 10-3 U/ml] and high selectivity with CEs. In biological imaging, probe CR could especially locate in the ER in HepG2 cells. After cells were treated with orilistat, CR succeeded in monitoring the changes in the CEs. Importantly, CR also had the ability to trace the changes in CEs in a tunicamycin-induced ER stress model. Therefore, probe CR could be a powerful molecular tool for further investigating the functions of CEs in the ER.

Semi-supervised classification of fundus images combined with CNN and GCN.

PURPOSE: Diabetic retinopathy (DR) is one of the most serious complications of diabetes, which is a kind of fundus lesion with specific changes. Early diagnosis of DR can effectively reduce the visual damage caused by DR. Due to the variety and different morphology of DR lesions, automatic classification of fundus images in mass screening can greatly save clinicians' diagnosis time. To alleviate these problems, in this paper, we propose a novel framework-graph attentional convolutional neural network (GACNN). METHODS AND MATERIALS: The network consists of convolutional neural network (CNN) and graph convolutional network (GCN). The global and spatial features of fundus images are extracted by using CNN and GCN, and attention mechanism is introduced to enhance the adaptability of GCN to topology map. We adopt semi-supervised method for classification, which greatly improves the generalization ability of the network. RESULTS: In order to verify the effectiveness of the network, we conducted comparative experiments and ablation experiments. We use confusion matrix, precision, recall, kappa score, and accuracy as evaluation indexes. With the increase of the labeling rates, the classification accuracy is higher. Particularly, when the labeling rate is set to 100%, the classification accuracy of GACNN reaches 93.35%. Compared with DenseNet121, the accuracy rate is improved by 6.24%. CONCLUSIONS: Semi-supervised classification based on attention mechanism can effectively improve the classification performance of the model, and attain preferable results in classification indexes such as accuracy and recall. GACNN provides a feasible classification scheme for fundus images, which effectively reduces the screening human resources.

Aggressive hydration compared to standard hydration with lactated ringer's solution for prevention of post endoscopic retrograde cholangiopancreatography pancreatitis.

BACKGROUND AND AIMS: Previous studies have suggested that aggressive hydration with lactated ringer solution are one of the protective factors in preventing post endoscopic retrograde cholangiopancreatography (post-ERCP). We conducted a systematic review and meta-analysis to examine the efficacy aggressive hydration with lactated Ringer solution in preventing PEP. METHODS: All published and unpublished articles on aggressive hydration with lactated ringer solution in those underwent ERCP procedure for any reasons were screened for eligibility. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed. This paper doesn't need the IRB approval. RESULTS: Seven RCTs met the inclusion criteria. Meta-analysis indicates that aggressive hydration with lactated Ringer solution were associated with lower PEP rate.[odds ratio (OR) 0.29; 95% confidence interval (CI) 0.18-0.48]; lower incidence of hyperamylasemia (OR 0.49; 95% CI 0.35, 0.69) and lower risk of pain (OR 0.28; 95% CI 0.10-0.81). The association between aggressive hydration with lactated Ringer solution and incidence of moderate severity PEP were unclear (OR 0.57; 95% CI 0.22, 1.45). Sensitivity analyses also showed that omitting 1 study from analysis of PEP rate could reduce the heterogeneity but did not change the conclusion of this meta-analysis. A cumulating meta-analysis was performed statistically which showed a stable result of overall incidence of PEP. CONCLUSIONS: Aggressive hydration with lactated Ringer solution was a protective factor in reducing the overall incidence of PEP, hyperamylasemia and risk of abdominal pain.

CT-Based Hand-crafted Radiomic Signatures Can Predict PD-L1 Expression Levels in Non-small Cell Lung Cancer: a Two-Center Study.

Here, we used pre-treatment CT images to develop and evaluate a radiomic signature that can predict the expression of programmed death ligand 1 (PD-L1) in non-small cell lung cancer (NSCLC). We then verified its predictive performance by cross-referencing its results with clinical characteristics. This two-center retrospective analysis included 125 patients with histologically confirmed NSCLC. A total of 1287 hand-crafted radiomic features were observed from manually determined tumor regions. Valuable features were then selected with a ridge regression-based recursive feature elimination approach. Machine learning-based prediction models were then built from this and compared each other. The final radiomic signature was built using logistic regression in the primary cohort, and then tested in a validation cohort. Finally, we compared the efficacy of the radiomic signature to the clinical model and the radiomic-clinical nomogram. Among the 125 patients, 89 were classified as having PD-L1 positive expression. However, there was no significant difference in PD-L1 expression levels determined by clinical characteristics (P = 0.109-0.955). Upon selecting 9 radiomic features, we found that the logistic regression-based prediction model performed the best (AUC = 0.96, P < 0.001). In the external cohort, our radiomic signature showed an AUC of 0.85, which outperformed both the clinical model (AUC = 0.38, P < 0.001) and the radiomics-nomogram model (AUC = 0.61, P < 0.001). Our CT-based hand-crafted radiomic signature model can effectively predict PD-L1 expression levels, providing a noninvasive means of better understanding PD-L1 expression in patients with NSCLC.

Non-coding RNAS and colorectal cancer liver metastasis.

More than 50% of colorectal cancer (CRC) deaths are attributed to metastasis, and the liver is the most common distant metastatic site of CRC. The molecular mechanisms underlying CRC liver metastasis are very complicated and remain largely unknown. Accumulated evidence has shown that non-coding RNAs (NcRNAs) play critical roles in tumor development and progression. Here we reviewed the roles and underlying mechanisms of NcRNAs in CRC liver metastasis.

Pre-treatment ADC image-based random forest classifier for identifying resistant rectal adenocarcinoma to neoadjuvant chemoradiotherapy.

OBJECTIVE: To develop a predicting model for tumor resistance to neoadjuvant chemoradiotherapy (NCRT) in locally advanced rectal cancer (LARC) by using pre-treatment apparent diffusion coefficient (ADC) image-derived radiomics features. METHOD: A total of 89 patients with LARC were randomly assigned into training (N = 66) and testing cohorts (N = 23) at the ratio of 3:1. Radiomics features were derived from manually determined tumor region of pre-treatment ADC images. Random forest algorithm was used to determine the most relevant features and then to construct a predicting model for identifying resistant tumor. Stability and diagnostic performance of the random forest model was evaluated with the testing cohort. RESULTS: The top 10 most relevant features (entropymean, inverse variance, energymean, small area emphasis, ADCmin, ADCmean, sdGa02, small gradient emphasis, age, and size) were determined from clinical characteristics and 133 radiomics features. In the prediction of resistant tumor of the testing cohort, the random forest model constructed based on these most relevant features achieved an area under the receiver operating characteristic curve of 0.83, with the highest accuracy of 91.3%, a sensitivity of 88.9%, and a specificity of 92.8%. CONCLUSION: The random forest classifier based on radiomics features derived from pre-treatment ADC images have the potential to predict tumor resistance to NCRT in patients with LARC, and the use of predicting model may facilitate individualized management of rectal cancer.

PIK3CD induces cell growth and invasion by activating AKT/GSK-3ß/ß-catenin signaling in colorectal cancer.

The catalytic subunit p110δ of phosphoinositide 3-kinase (PI3K) encoded by PIK3CD has been implicated in some human solid tumors. However, its roles in colorectal cancer (CRC) remain largely unknown. Here we found that PIK3CD was overexpressed in colon cancer tissues and CRC cell lines and was an independent predictor for overall survival (OS) of patients with colon cancer. The ectopic overexpression of PIK3CD significantly promoted CRC cell growth, migration and invasion in vitro and tumor growth in vivo. In contrast, inhibition of PIK3CD by specific small-interfering RNA or idelalisib dramatically suppressed CRC cell growth, migration and invasion in vitro and tumor growth in vivo. Moreover, PIK3CD overexpression increased AKT activity, nuclear translocation of ß-catenin and T-cell factor/lymphoid enhancer factor (TCF/LEF) transcriptional activity and decreased glycogen synthase kinase 3ß (GSK-3ß) activity, whereas PIK3CD inhibition exhibited the opposite effects. Furthermore, PIK3CD-mediated cell growth, migration and invasion were reversed by blockade of AKT signaling or depletion of ß-catenin. In addition, PIK3CD expression in colon cancer tissues positively correlated with ß-catenin abnormal expression, which was an independent predictor for OS of colon cancer patients. Taken together, our findings demonstrate that PIK3CD is an independent prognostic factor in CRC and that PIK3CD induces CRC cell growth, migration and invasion by activating AKT/GSK-3ß/ß-catenin signaling, suggesting that PIK3CD might be a novel prognostic biomarker and a potential therapeutic target for CRC.

Development of a polarity-sensitive ratiometric fluorescent probe based on the intramolecular reaction of spiro-oxazolidine and its applications for in situ visualizing the fluctuations of polarity during ER stress.

Polarity is a vital element in endoplasmic reticulum (ER) microenvironment, and its variation is closely related to many physiological and pathological activities of ER, so it is necessary to trace fluctuations of polarity in ER. However, most of fluorescent probes for detecting polarity dependent on the changes of single emission, which could be affected by many factors and cause false signals. Ratiometric fluorescent probe with "built-in calibration" can effectively avoid detection errors. Here, we have designed a ratiometric fluorescent probe HM for monitoring the ER polarity based on the intramolecular reaction of spiro-oxazolidine. It forms ring open/closed isomers driven by polarity to afford ratiometric sensing. Probe HM have manifested its ratiometric responses to polarity in spectroscopic results, which could offer much more precise information for the changes of polarity in living cells with the internal built-in correction. It also showed large emission shift ( 133 nm), high selectivity and photo-stability. In biological imaging, HM could selectively accumulate in ER with high photo-stability. Importantly, HM has ability for in situ tracing the changes of ER polarity with ratiometric behavior during the ER stress process with the stimulation of tunicamycin, dithiothreitol and hypoxia, suggesting that HM is an effective molecule tool for monitoring the variations of ER polarity.

A pH-response-based fluorescent probe for detecting the mitophagy process by tracing changes in colocalization coefficients.

Mitochondria are not only the center of energy metabolism but also involved in regulating cellular activities. Quality and quantity control of mitochondria is therefore essential. Mitophagy is a lysosome-dependent process to clear dysfunctional mitochondria, and abnormal mitophagy can cause metabolic disorders. Therefore, it is necessary to monitor the mitophagy in living cells on a real-time basis. Herein, we developed a pH-responsive fluorescent probe MP for the detection of the mitophagy process using real-time tracing colocalization coefficients. Probe MP showed good pH responses with high selectivity and sensitivity in spectral testing. Probe MP is of positive charge, which is beneficial for accumulating into mitochondrial in living cells. Cells exhibited pH-dependent fluorescence when they were treated with different pH media. Importantly, the changes in the colocalization coefficient between probe MP and Lyso Tracker® Deep Red from 0.4 to 0.8 were achieved in a real-time manner during the mitophagy stimulated by CCCP, starvation and rapamycin. Therefore, combined with the parameter of the colocalization coefficient, probe MP is a potential molecular tool for the real-time tracing of mitophagy to further explore the details of mitophagy.

Hierarchical-Instance Contrastive Learning for Minority Detection on Imbalanced Medical Datasets.

Deep learning methods are often hampered by issues such as data imbalance and data-hungry. In medical imaging, malignant or rare diseases are frequently of minority classes in the dataset, featured by diversified distribution. Besides that, insufficient labels and unseen cases also present conundrums for training on the minority classes. To confront the stated problems, we propose a novel Hierarchical-instance Contrastive Learning (HCLe) method for minority detection by only involving data from the majority class in the training stage. To tackle inconsistent intra-class distribution in majority classes, our method introduces two branches, where the first branch employs an auto-encoder network augmented with three constraint functions to effectively extract image-level features, and the second branch designs a novel contrastive learning network by taking into account the consistency of features among hierarchical samples from majority classes. The proposed method is further refined with a diverse mini-batch strategy, enabling the identification of minority classes under multiple conditions. Extensive experiments have been conducted to evaluate the proposed method on three datasets of different diseases and modalities. The experimental results show that the proposed method outperforms the state-of-the-art methods.

Sparse-view cone-beam computed tomography iterative reconstruction based on new multi-gradient direction total variation.

AIM: The accurate reconstruction of cone-beam computed tomography (CBCT) from sparse projections is one of the most important areas for study. The compressed sensing theory has been widely employed in the sparse reconstruction of CBCT. However, the total variation (TV) approach solely uses information from the i-coordinate, j-coordinate, and k-coordinate gradients to reconstruct the CBCT image. MATERIALS AND METHODS: It is well recognized that the CBCT image can be reconstructed more accurately with more gradient information from different directions. Thus, this study introduces a novel approach, named the new multi-gradient direction total variation minimization method. The method uses gradient information from the ij-coordinate, ik-coordinate, and jk-coordinate directions to reconstruct CBCT images, which incorporates nine different types of gradient information from nine directions. RESULTS: This study assessed the efficacy of the proposed methodology using under-sampled projections from four different experiments, including two digital phantoms, one patient's head dataset, and one physical phantom dataset. The results indicated that the proposed method achieved the lowest RMSE index and the highest SSIM index. Meanwhile, we compared the voxel intensity curves of the reconstructed images to assess the edge structure preservation. Among the various methods compared, the curves generated by the proposed method exhibited the highest level of consistency with the gold standard image curves. CONCLUSION: In summary, the proposed method showed significant potential in enhancing the quality and accuracy of CBCT image reconstruction.

Ultrasound deep learning radiomics and clinical machine learning models to predict low nuclear grade, ER, PR, and HER2 receptor status in pure ductal carcinoma in situ.

Background: Low nuclear grade ductal carcinoma in situ (DCIS) patients can adopt proactive management strategies to avoid unnecessary surgical resection. Different personalized treatment modalities may be selected based on the expression status of molecular markers, which is also predictive of different outcomes and risks of recurrence. DCIS ultrasound findings are mostly non mass lesions, making it difficult to determine boundaries. Currently, studies have shown that models based on deep learning radiomics (DLR) have advantages in automatic recognition of tumor contours. Machine learning models based on clinical imaging features can explain the importance of imaging features. Methods: The available ultrasound data of 349 patients with pure DCIS confirmed by surgical pathology [54 low nuclear grade, 175 positive estrogen receptor (ER+), 163 positive progesterone receptor (PR+), and 81 positive human epidermal growth factor receptor 2 (HER2+)] were collected. Radiologists extracted ultrasonographic features of DCIS lesions based on the 5th Edition of Breast Imaging Reporting and Data System (BI-RADS). Patient age and BI-RADS characteristics were used to construct clinical machine learning (CML) models. The RadImageNet pretrained network was used for extracting radiomics features and as an input for DLR modeling. For training and validation datasets, 80% and 20% of the data, respectively, were used. Logistic regression (LR), support vector machine (SVM), random forest (RF), and eXtreme Gradient Boosting (XGBoost) algorithms were performed and compared for the final classification modeling. Each task used the area under the receiver operating characteristic curve (AUC) to evaluate the effectiveness of DLR and CML models. Results: In the training dataset, low nuclear grade, ER+, PR+, and HER2+ DCIS lesions accounted for 19.20%, 65.12%, 61.21%, and 30.19%, respectively; the validation set, they consisted of 19.30%, 62.50%, 57.14%, and 30.91%, respectively. In the DLR models we developed, the best AUC values for identifying features were 0.633 for identifying low nuclear grade, completed by the XGBoost Classifier of ResNet50; 0.618 for identifying ER, completed by the RF Classifier of InceptionV3; 0.755 for identifying PR, completed by the XGBoost Classifier of InceptionV3; and 0.713 for identifying HER2, completed by the LR Classifier of ResNet50. The CML models had better performance than DLR in predicting low nuclear grade, ER+, PR+, and HER2+ DCIS lesions. The best AUC values by classification were as follows: for low nuclear grade by RF classification, AUC: 0.719; for ER+ by XGBoost classification, AUC: 0.761; for PR+ by XGBoost classification, AUC: 0.780; and for HER2+ by RF classification, AUC: 0.723. Conclusions: Based on small-scale datasets, our study showed that the DLR models developed using RadImageNet pretrained network and CML models may help predict low nuclear grade, ER+, PR+, and HER2+ DCIS lesions so that patients benefit from hierarchical and personalized treatment.

A difficulty-aware and task-augmentation method based on meta-learning model for few-shot diabetic retinopathy classification.

Background: Accurate classification techniques are essential for the early diagnosis and treatment of patients with diabetic retinopathy (DR). However, the limited amount of annotated DR data poses a challenge for existing deep-learning models. This article proposes a difficulty-aware and task-augmentation method based on meta-learning (DaTa-ML) model for few-shot DR classification with fundus images. Methods: The difficulty-aware (Da) method operates by dynamically modifying the cross-entropy loss function applied to learning tasks. This methodology has the ability to intelligently down-weight simpler tasks, while simultaneously prioritizing more challenging tasks. These adjustments occur automatically and aim to optimize the learning process. Additionally, the task-augmentation (Ta) method is used to enhance the meta-training process by augmenting the number of tasks through image rotation and improving the feature-extraction capability. To implement the expansion of the meta-training tasks, various task instances can be sampled during the meta-training stage. Ultimately, the proposed Ta method was introduced to optimize the initialization parameters and enhance the meta-generalization performance of the model. The DaTa-ML model showed promising results by effectively addressing the challenges associated with few-shot DR classification. Results: The Asia Pacific Tele-Ophthalmology Society (APTOS) 2019 blindness detection data set was used to evaluate the DaTa-ML model. The results showed that with only 1% of the training data (5-way, 20-shot) and a single update step (training time reduced by 90%), the DaTa-ML model had an accuracy rate of 89.6% on the test data, which is a 1.7% improvement over the transfer-learning method [i.e., residual neural network (ResNet)50 pre-trained on ImageNet], and a 16.8% improvement over scratch-built models (i.e., ResNet50 without pre-trained weights), despite having fewer trainable parameters (the parameters used by the DaTa-ML model are only 0.47% of the ResNet50 parameters). Conclusions: The DaTa-ML model provides a more efficient DR classification solution with little annotated data and has significant advantages over state-of-the-art methods. Thus, it could be used to guide and assist ophthalmologists to determine the severity of DR.