Combination of prostate volume and apparent diffusion coefficient can stratify patients with a PI-RADS score of 3 to reduce unnecessary prostate biopsies.

BACKGROUND: Nowadays, there are many patients who undergo unnecessary prostate biopsies after receiving a prostate imaging reporting and data system (PI-RADS) score of 3. Our purpose is to identify cutoff values of the prostate volume (PV) and minimum apparent diffusion coefficient (ADCmin) to stratify those patients to reduce unnecessary prostate biopsies. METHODS: Data from 224 qualified patients who received prostate biopsies from January 2019 to June 2023 were collected. The Mann-Whitney U test was used to compare non-normal distributed continuous variables, which were recorded as median (interquartile ranges). The correlation coefficients were calculated using Spearman's rank correlation analysis. Categorical variables are recorded by numbers (percentages) and compared by χ2 test. Both univariate and multivariate logistic regression analysis were used to determine the independent predictors. The receiver-operating characteristic curve and the area under the curve (AUC) were used to evaluate the diagnostic performance of clinical variables. RESULTS: Out of a total of 224 patients, 36 patients (16.07%) were diagnosed with clinically significant prostate cancer (csPCa), whereas 72 patients (32.14%) were diagnosed with any grade prostate cancer. The result of multivariate analysis demonstrated that the PV (p < 0.001, odds ratio [OR]: 0.952, 95% confidence interval [95% CI]: 0.927-0.978) and ADCmin (p < 0.01, OR: 0.993, 95% CI: 0.989-0.998) were the independent factors for predicting csPCa. The AUC values of the PV and ADCmin were 0.779 (95% CI: 0.718-0.831) and 0.799 (95% CI: 0.740-0.849), respectively, for diagnosing csPCa. After stratifying patients by PV and ADCmin, 24 patients (47.06%) with "PV < 55 mL and ADCmin < 685 µm2/s" were diagnosed with csPCa. However, only one patient (1.25%) with PV ≥ 55 mL and ADCmin ≥ 685 µm2/s were diagnosed with csPCa. CONCLUSIONS: In this study, we found the combination of PV and ADCmin can stratify patients with a PI-RADS score of 3 to reduce unnecessary prostate biopsies. These patients with "PV ≥ 55 mL and ADCmin ≥ 685 µm2/s" may safely avoid prostate biopsies.

Creating a novel multiparametric magnetic resonance imaging-based biopsy strategy for reducing unnecessary prostate biopsies: a retrospective cohort study.

Background: The overdiagnosis of prostate cancer (PCa) caused by unnecessary prostate biopsy has become a worldwide problem that urgently requires a solution. We aimed to reduce the unnecessary prostate biopsies and increase the detection rate of clinically significant PCa (csPCa) by creating a novel multiparametric magnetic resonance imaging (mpMRI)-based strategy. Methods: A total of 1,194 eligible patients who underwent transperineal prostate biopsies from January 2018 to December 2022 were included in this retrospective study. Of these patients, 1,080 who received prostate biopsies from January 2018 to July 2022 were regarded as cohort 1 for primary analysis, and 114 patients who received prostate biopsies from August 2022 to December 2022 were collected in cohort 2 for validation. All the mpMRI images were quantitatively evaluated by the Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v. 2.1). The diagnostic performances were assessed through the receiver operating characteristic (ROC) curve and area under the curve (AUC) and were compared with the DeLong test. Cancer diagnosis-free survival analysis was performed using the Kaplan-Meier method and log-rank test. The primary endpoint of this study was clinically significant PCa with an International Society of Urological Pathology (ISUP) grade ≥2. Results: In cohort 1, the results of ROC curves demonstrated that the PI-RADS score had a higher diagnostic accuracy (AUC =0.898 for any-grade PCa; AUC =0.917 for csPCa) than did the other clinical variables (P<0.001). Under the novel mpMRI-based biopsy strategy, all patients with PI-RADS 1 can safely avoid prostate biopsy. For patients with PI-RADS 2, prostate biopsy should be considered for patients with prostate-specific antigen density (PSAD) ≥0.3 ng/mL2 and prostate volume <65 mL. As for patients with PI-RADS 3, structured surveillance programs can be a viable option if PSAD <0.3 ng/mL2 and prostate volume ≥65 mL. Finally, patients with a PI-RADS score of 4 and 5 should undergo prostate biopsy due to the high probability of clinically significant PCa. In the validation analysis of cohort 2, 48 patients were placed into a biopsy-spared group with no csPCa cases, while 66 patients were placed in a biopsy-needed group, with an csPCa detection rate of 50.0%. Overall, the novel strategy demonstrated a sensitivity, specificity, positive predictive value, and negative predictive value of 98.9%, 57.5%, 50.5%, and 99.2%, respectively, for diagnosing csPCa. Conclusions: An mpMRI-based biopsy strategy can effectively avoid about 40% of prostate biopsies and maintain a high detection rate for clinically significant PCa. It can further provide valuable guidance for patients and physicians in considering the necessity of prostate biopsy.

Protocol for SNOTOB study: radical prostatectomy without prostate biopsy following 18F-PSMA-1007 PET/CT based on a diagnostic model: a single-centre, single-arm, open-label study.

INTRODUCTION: Nowadays, invasive prostate biopsy is the standard diagnostic test for patients with suspected prostate cancer (PCa). However, it has some shortcomings such as perioperative complications, economic and psychological burden on patients, and some patients may undergo repeated prostate biopsy. In this study protocol, our aim is to provide a non-invasive diagnostic strategy we call the 'prostate-specific membrane antigen (PSMA) combined model' for the diagnosis of PCa. If patients are diagnosed with PCa using PSMA combined model, we want to prove these patients can receive radical prostatectomy directly without prior prostate biopsies. METHODS: The SNOTOB trial adopts a prospective, single-centre, single-arm, open-label study design. The PSMA combined model consists of a diagnostic model based on what we previously reported and 18F-PSMA-1007 positron emission tomography/CT (18F-PSMA-1007 PET/CT) examinations in series. First, patients use the diagnostic model (online address: https://ustcprostatecancerprediction.shinyapps.io/dynnomapp/) to calculate the risk probability of clinically significant PCa (csPCa). When the risk probability of csPCa is equal or greater than 0.60, 18F-PSMA-1007 PET/CT will be applied for further diagnosis. If patients are still considered as csPCa after 18F-PSMA-1007 PET/CT examinations, we define this condition as positive results of PSMA combined model. Subsequently, we will recommend these patients to accept radical prostatectomy without prostate biopsy directly. Finally, the diagnostic performance of PSMA combined model will be verified with the pathological results. Totally, 57 patients need to be enrolled in this clinical trial. ETHICS AND DISSEMINATION: This study was approved by the ethics committee of The First Affiliated Hospital of USTC (No. 2022KY-142). The results of this study will be published in peer-reviewed journals and reported at academic conferences. TRIAL REGISTRATION NUMBER: NCT05587192.

Developing a diagnostic model for predicting prostate cancer: a retrospective study based on Chinese multicenter clinical data.

ABSTRACT: The overdiagnosis of prostate cancer (PCa) caused by nonspecific elevation serum prostate-specific antigen (PSA) and the overtreatment of indolent PCa have become a global problem that needs to be solved urgently. We aimed to construct a prediction model and provide a risk stratification system to reduce unnecessary biopsies. In this retrospective study, clinical data of 1807 patients from three Chinese hospitals were used. The final model was built using stepwise logistic regression analysis. The apparent performance of the model was assessed by receiver operating characteristic curves, calibration plots, and decision curve analysis. Finally, a risk stratification system of clinically significant prostate cancer (csPCa) was created, and diagnosis-free survival analyses were performed. Following multivariable screening and evaluation of the diagnostic performances, a final diagnostic model comprised of the PSA density and Prostate Imaging-Reporting and Data System (PI-RADS) score was established. Model validation in the development cohort and two external cohorts showed excellent discrimination and calibration. Finally, we created a risk stratification system using risk thresholds of 0.05 and 0.60 as the cut-off values. The follow-up results indicated that the diagnosis-free survival rate for csPCa at 12 months and 24 months postoperatively was 99.7% and 99.4%, respectively, for patients with a risk threshold below 0.05 after the initial negative prostate biopsy, which was significantly better than patients with higher risk. Our diagnostic model and risk stratification system can achieve a personalized risk calculation of csPCa. It provides a standardized tool for Chinese patients and physicians when considering the necessity of prostate biopsy.

Generalizability of Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs Using an Adaptation of the Modified-2 Algorithm-Based Qualitative Criteria.

RATIONALE AND OBJECTIVES: Spinal osteoporotic compression fractures (OCFs) can be an early biomarker for osteoporosis but are often subtle, incidental, and underreported. To ensure early diagnosis and treatment of osteoporosis, we aimed to build a deep learning vertebral body classifier for OCFs as a critical component of our future automated opportunistic screening tool. MATERIALS AND METHODS: We retrospectively assembled a local dataset, including 1790 subjects and 15,050 vertebral bodies (thoracic and lumbar). Each vertebral body was annotated using an adaption of the modified-2 algorithm-based qualitative criteria. The Osteoporotic Fractures in Men (MrOS) Study dataset provided thoracic and lumbar spine radiographs of 5994 men from six clinical centers. Using both datasets, five deep learning algorithms were trained to classify each individual vertebral body of the spine radiographs. Classification performance was compared for these models using multiple metrics, including the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, and positive predictive value (PPV). RESULTS: Our best model, built with ensemble averaging, achieved an AUC-ROC of 0.948 and 0.936 on the local dataset's test set and the MrOS dataset's test set, respectively. After setting the cutoff threshold to prioritize PPV, this model achieved a sensitivity of 54.5% and 47.8%, a specificity of 99.7% and 99.6%, and a PPV of 89.8% and 94.8%. CONCLUSION: Our model achieved an AUC-ROC>0.90 on both datasets. This testing shows some generalizability to real-world clinical datasets and a suitable performance for a future opportunistic osteoporosis screening tool.

Mechanisms of circular RNA degradation.

Circular RNAs (CircRNAs) are a class of noncoding RNAs formed by backsplicing during cotranscriptional and posttranscriptional processes, and they widely exist in various organisms. CircRNAs have multiple biological functions and are associated with the occurrence and development of many diseases. While the biogenesis and biological function of circRNAs have been extensively studied, there are few studies on circRNA degradation and only a few pathways for specific circRNA degradation have been identified. Here we outline basic information about circRNAs, summarize the research on the circRNA degradation mechanisms and discusses where this field might head, hoping to provide some inspiration and guidance for scholars who aim to study the degradation of circRNAs.

Improving the Accuracy of Progress Indication for Constructing Deep Learning Models.

For many machine learning tasks, deep learning greatly outperforms all other existing learning algorithms. However, constructing a deep learning model on a big data set often takes days or months. During this long process, it is preferable to provide a progress indicator that keeps predicting the model construction time left and the percentage of model construction work done. Recently, we developed the first method to do this that permits early stopping. That method revises its predicted model construction cost using information gathered at the validation points, where the model's error rate is computed on the validation set. Due to the sparsity of validation points, the resulting progress indicators often have a long delay in gathering information from enough validation points and obtaining relatively accurate progress estimates. In this paper, we propose a new progress indication method to overcome this shortcoming by judiciously inserting extra validation points between the original validation points. We implemented this new method in TensorFlow. Our experiments show that compared with using our prior method, using this new method reduces the progress indicator's prediction error of the model construction time left by 57.5% on average. Also, with a low overhead, this new method enables us to obtain relatively accurate progress estimates faster.

Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs using an Adaptation of the Genant Semiquantitative Criteria.

RATIONALE AND OBJECTIVES: Osteoporosis affects 9% of individuals over 50 in the United States and 200 million women globally. Spinal osteoporotic compression fractures (OCFs), an osteoporosis biomarker, are often incidental and under-reported. Accurate automated opportunistic OCF screening can increase the diagnosis rate and ensure adequate treatment. We aimed to develop a deep learning classifier for OCFs, a critical component of our future automated opportunistic screening tool. MATERIALS AND METHODS: The dataset from the Osteoporotic Fractures in Men Study comprised 4461 subjects and 15,524 spine radiographs. This dataset was split by subject: 76.5% training, 8.5% validation, and 15% testing. From the radiographs, 100,409 vertebral bodies were extracted, each assigned one of two labels adapted from the Genant semiquantitative system: moderate to severe fracture vs. normal/trace/mild fracture. GoogLeNet, a deep learning model, was trained to classify the vertebral bodies. The classification threshold on the predicted probability of OCF outputted by GoogLeNet was set to prioritize the positive predictive value (PPV) while balancing it with the sensitivity. Vertebral bodies with the top 0.75% predicted probabilities were classified as moderate to severe fracture. RESULTS: Our model yielded a sensitivity of 59.8%, a PPV of 91.2%, and an F1 score of 0.72. The areas under the receiver operating characteristic curve (AUC-ROC) and the precision-recall curve were 0.99 and 0.82, respectively. CONCLUSION: Our model classified vertebral bodies with an AUC-ROC of 0.99, providing a critical component for our future automated opportunistic screening tool. This could lead to earlier detection and treatment of OCFs.

DicomAnnotator: a Configurable Open-Source Software Program for Efficient DICOM Image Annotation.

Modern, supervised machine learning approaches to medical image classification, image segmentation, and object detection usually require many annotated images. As manual annotation is usually labor-intensive and time-consuming, a well-designed software program can aid and expedite the annotation process. Ideally, this program should be configurable for various annotation tasks, enable efficient placement of several types of annotations on an image or a region of an image, attribute annotations to individual annotators, and be able to display Digital Imaging and Communications in Medicine (DICOM)-formatted images. No current open-source software program fulfills these requirements. To fill this gap, we developed DicomAnnotator, a configurable open-source software program for DICOM image annotation. This program fulfills the above requirements and provides user-friendly features to aid the annotation process. In this paper, we present the design and implementation of DicomAnnotator. Using spine image annotation as a test case, our evaluation showed that annotators with various backgrounds can use DicomAnnotator to annotate DICOM images efficiently. DicomAnnotator is freely available at https://github.com/UW-CLEAR-Center/DICOM-Annotator under the GPLv3 license.

Progress Indication for Deep Learning Model Training: A Feasibility Demonstration.

Deep learning is the state-of-the-art learning algorithm for many machine learning tasks. Yet, training a deep learning model on a large data set is often time-consuming, taking several days or even months. During model training, it is desirable to offer a non-trivial progress indicator that can continuously project the remaining model training time and the fraction of model training work completed. This makes the training process more user-friendly. In addition, we can use the information given by the progress indicator to assist in workload management. In this paper, we present the first set of techniques to support non-trivial progress indicators for deep learning model training when early stopping is allowed. We report an implementation of these techniques in TensorFlow and our evaluation results for both convolutional and recurrent neural networks. Our experiments show that our progress indicator can offer useful information even if the run-time system load varies over time. In addition, the progress indicator can self-correct its initial estimation errors, if any, over time.

[Influence of Cibotium barametz and its processed samples on haemorheology index in mice with adjuvant arthritis].

OBJECTIVE: To discuss the influence of the rhizome of Cibotium barametz on the heamorheology index in mice with adjuvant arthritis and to compare the effect of raw medicinals with that of the processed ones. METHOD: Mice was injected with Freund's complete adjuvant on the rihgt behind foot to make model of adjuvant arthritis (AA). Hydroxyacrbamide tablets were orally administrated by mice with AA to make model of AA due to deficiency in the kidney (DK-AA). And then we determined the heamorheology index of the normal group, positive control group, AA group, DK-AA group and medicinals-treated groups. RESULT: In the groups of AA, and DK-AA, the heamorheology index, such as high shearing, middle shearing, low shearing, plasma viscosity, whole blood reduction viscosity, erythrocyte aggregation exponent, erythrocyte degeneration exponent, sedimentation, sedimentation equation K value, erythrocyte rigidity exponent, erythrocyte electrophoresis time, casson viscosity, casson yield stress, increased significantly. After treated with Cibotium barametz, the heamorheology index except red blood count, packed cell volume, fibrinogen decreased obviously to get normal. CONCLUSION: Rhizome of Cibotium barametz could promote heamorheology in mice with AA and DK-AA to exhibit effect of promoting blood circulation and remove blood stasis. The medicinal rhizomes processed with sand have the effect enhanced.