相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Oncology
Wei Wang et al.
Summary: The purpose of this study was to develop predictive models based on the integration of radiomics with VIRADS for determining muscle invasion of bladder cancer. The results showed that both the radiomics model and the clinical-radiomics model had good performance in the diagnosis of bladder cancer, which was especially useful for less experienced readers.
UROLOGIC ONCOLOGY-SEMINARS AND ORIGINAL INVESTIGATIONS
(2023)
Article
Multidisciplinary Sciences
Yusaku Moribata et al.
Summary: This study developed a versatile automatic segmentation model for bladder cancer (BC) using a convolutional neural network and investigated the robustness of radiomics features extracted from ADC maps. The model achieved high accuracy in BC segmentation and the extracted radiomics features exhibited high reproducibility. The study highlights the potential of using a deep learning model for automatic BC segmentation and the reliability of radiomics features for cancer diagnosis.
SCIENTIFIC REPORTS
(2023)
Review
Radiology, Nuclear Medicine & Medical Imaging
Olivier Rouviere et al.
Summary: The purpose of this study was to review the diagnostic performance of AI-based algorithms for characterizing/detecting prostate cancer on MRI. The algorithms showed promising results and performed well as standalone diagnostic tools. However, the management of discrepancies between algorithm findings and human reading needs further investigation.
DIAGNOSTIC AND INTERVENTIONAL IMAGING
(2023)
Article
Oncology
Xinxin Yu et al.
Summary: A retrospective study developed a mathematical model combining radiomics and clinical factors to predict synchronous distant metastasis in patients with clear cell renal cell carcinoma. The model showed high accuracy and clinical usefulness in the external validation group.
FRONTIERS IN ONCOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Xuhui Fan et al.
Summary: This study utilized radiomics to predict Ki67 and histological grade of bladder cancer preoperatively based on multiparameter MRI sequences, facilitating clinical decision-making. The results demonstrated the potential of radiomics to predict various pathological outcomes of bladder cancer and provide guidance for clinical decision-making.
BRITISH JOURNAL OF RADIOLOGY
(2023)
Article
Computer Science, Interdisciplinary Applications
Jianpeng Li et al.
Summary: This study compared the effectiveness of radiomics, single-task deep learning (DL), and multi-task DL methods in predicting muscle-invasive bladder cancer (MIBC) status based on T2-weighted imaging (T2WI). The multi-task model performed better in the test cohort compared to other models. The multi-task DL method had the advantage of saving time and effort compared to radiomics, and it was more lesion-focused and reliable for clinical reference compared to the single-task DL method.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Nityanand Miskin et al.
Summary: This study aimed to differentiate benign from malignant cystic renal masses using artificial intelligence techniques. The results showed that the accuracy of identifying benign and malignant masses ranged from 0.61 to 0.79 for different models, indicating the potential of computed tomography texture-based machine learning algorithms as an adjunct to radiologists' assessments.
JOURNAL OF COMPUTER ASSISTED TOMOGRAPHY
(2023)
Article
Urology & Nephrology
Eva Gresser et al.
Summary: This study aimed to evaluate the performance of quantitative radiomics signatures in the detection of lymph node metastases in bladder cancer. By analyzing 1354 patients who underwent radical cystectomy, it was found that the diagnostic efficacy of radiologist-evaluated CT imaging studies is limited. The results showed that radiomics signatures can accurately discriminate lymph node metastases.
EUROPEAN UROLOGY FOCUS
(2023)
Article
Multidisciplinary Sciences
Ana Rodrigues et al.
Summary: There is evidence that artificial intelligence may be helpful in prostate cancer. However, building robust machine learning algorithms that account for radiologist segmentation variability is a challenge.
SCIENTIFIC REPORTS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Pengfei Jin et al.
Summary: The purpose of this study was to develop machine learning-based radiomics models derived from different MRI sequences to distinguish between benign and malignant PI-RADS 3 lesions before intervention, and to validate the generalization ability of the models across different institutions. The pre-biopsy MRI data of 463 patients classified as PI-RADS 3 lesions were retrospectively collected from 4 medical institutions. Radiomics features were extracted from the T2WI, DWI, and ADC images, and single-sequence models and an integrated model were constructed using ANOVA feature ranking and support vector machine classifier. The models were tested and validated internally and externally, and the predictive performance was compared using AUC.
BMC MEDICAL IMAGING
(2023)
Article
Oncology
Suryadipto Sarkar et al.
Summary: Accurate bladder cancer staging is crucial for determining appropriate treatment and improving patient outcome. This study presents a hybrid machine/deep learning model that improves the accuracy of bladder cancer staging using CT scans, facilitating clinical management and improving patient outcome.
Article
Oncology
Hye Won Lee et al.
Summary: By leveraging recent advances in deep learning, our study successfully predicted biochemical recurrence after prostate surgery using multiparametric magnetic resonance imaging combined with clinical variables. Our method outperformed existing methods, providing better patient care through individualized medical imaging.
Review
Urology & Nephrology
Matteo Ferro et al.
Summary: Radiomics and artificial intelligence (AI) have the potential to improve the differentiation and characterization of various renal lesions, including distinguishing benign from malignant lesions, differentiating different subtypes of renal cell carcinoma, and predicting clinical outcomes such as Fuhrman grade and treatment response. By analyzing imaging data using neural networks, quantitative features related to contour, heterogeneity, and gray zone can be extracted, providing valuable information for diagnosis and prognosis. Standardization of scanner protocols is crucial for enhancing the diagnostic ability of imaging tools and improving the accuracy in differentiating between benign and clinically significant renal cancers.
THERAPEUTIC ADVANCES IN UROLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Dominik Deniffel et al.
Summary: Adding CT-based radiomics to established clinico-pathological biomarkers improves recurrence risk prediction for adjuvant treatment decisions in patients with non-metastatic renal cell carcinoma.
EUROPEAN RADIOLOGY
(2023)
Article
Oncology
Francesco Prata et al.
Summary: The aim of the study was to develop a radiomic tool for predicting clinically significant prostate cancer. Radiomic analysis allowed for the prediction of a Gleason score >= 7, improving the detection rate of clinically significant prostate cancer and reducing the number of unnecessary biopsies by overcoming the limitations of subjective interpretation of magnetic resonance imaging.
Article
Radiology, Nuclear Medicine & Medical Imaging
Andrew L. Wentland et al.
Summary: A machine learning model trained with CT-based radiomics features can provide superior accuracy for distinguishing benign from malignant solid renal masses compared to abdominal radiologists.
ABDOMINAL RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ghazaleh Jamshidi et al.
Summary: This study evaluated radiomics models based on T2-W and DCE MRI for the diagnosis of prostate cancer. The results showed that T2-W images were more effective in PCa detection compared to DCE images, and a fusion model was proposed to improve accuracy.
MAGNETIC RESONANCE MATERIALS IN PHYSICS BIOLOGY AND MEDICINE
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Xiaoli Li et al.
Summary: The study developed a CT-based radiomics nomogram integrating clinical factors and radiomics features to differentiate between renal oncocytoma and chromophobe renal cell carcinoma with high diagnostic efficacy. The radiomics nomogram performed better than clinical models and radiomics signatures, showing great potential in improving differential diagnosis and reducing unnecessary surgeries for renal tumors.
BRITISH JOURNAL OF RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Zhilong Yi et al.
Summary: The study investigated the use of machine learning-based radiomics models derived from (68) Ga-PSMA PET images to predict invisible intraprostatic lesions. Results showed that these models could accurately predict such lesions and outperformed PSA density in diagnostic performance.
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Natalie L. Demirjian et al.
Summary: This study evaluated the utility of CT-based radiomics signatures in discriminating low-grade and high-grade, as well as low-stage and high-stage clear cell renal cell carcinomas (ccRCC). The results suggest that CT-based radiomics signatures have the potential to be valuable in distinguishing between different grades and stages of ccRCC.
EUROPEAN RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Gumuyang Zhang et al.
Summary: This study investigated the feasibility of a CT-based radiomics prediction model to evaluate muscle invasiveness in bladder cancer. The study found that the prediction model had good diagnostic performance in diagnosing muscle-invasive bladder cancer.
EUROPEAN RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Sehnaz Evrimler et al.
Summary: Histological variant of bladder urothelial carcinoma is an important factor for therapy management. This study aims to assess the predictive performance of machine learning-based Computed Tomography radiomics in predicting histological variants. The results suggest that this method has high predictive performance and can provide valuable information for treatment management.
ACADEMIC RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Haoxin Zheng et al.
Summary: This study retrospectively reviewed 330 patients with negative prostate mpMRI results and used a machine learning approach to predict prostate biopsy results. The model, utilizing radiomics and clinical features, demonstrated high performance in predicting negative prostate biopsy results, helping to stratify patients who may avoid prostate biopsy.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2022)
Article
Oncology
Xinxin Zhang et al.
Summary: This study developed an MRI-based radiomics nomogram to predict the response of MIBC patients to NAC. The results showed that the radiomics nomogram had good performance in predicting tumor response to NAC.
FRONTIERS IN ONCOLOGY
(2022)
Article
Oncology
Rakesh Shiradkar et al.
Summary: The combination of shape distension descriptors derived from T2-weighted MRI and texture radiomics of prostate cancer can improve the prediction of biochemical recurrence post-radical prostatectomy.
FRONTIERS IN ONCOLOGY
(2022)
Review
Medicine, General & Internal
Henrik J. Michaely et al.
Summary: The combination of biparametric prostate MRI with machine-learning and deep learning methods shows promising results in the detection, grading, and characterization of prostate cancer.
Review
Radiology, Nuclear Medicine & Medical Imaging
Nikita Sushentsev et al.
Summary: This study systematically reviewed the current literature to evaluate the ability of fully-automated deep learning (DL) and semi-automated traditional machine learning (TML) MRI-based artificial intelligence (AI) methods in differentiating clinically significant prostate cancer from indolent prostate cancer and benign conditions. The study found comparable performance of the two classes of AI methods and identified common methodological limitations and biases.
INSIGHTS INTO IMAGING
(2022)
Article
Medicine, General & Internal
Jonas Hugosson et al.
Summary: MRI-directed targeted biopsy for prostate cancer screening reduces the risk of overdiagnosis by half compared to systematic biopsy, but may delay the detection of intermediate-risk tumors in a small proportion of patients.
NEW ENGLAND JOURNAL OF MEDICINE
(2022)
Article
Oncology
Andrew M. Fang et al.
Summary: Elevated PSA density, older age, and a biopsy-naive status are associated with clinically significant prostate cancer (CSPCa) in men with PI-RADS 3 lesions, while a prior negative biopsy is negatively associated. A predictive model can prevent unnecessary biopsies for PI-RADS 3 lesions while missing only a few cases of CSPCa.
Article
Oncology
Piotr Woznicki et al.
Summary: This study investigates the potential of quantitative features from preoperative CT images to predict overall survival in bladder cancer patients treated with radical cystectomy. Regression models based on radiomics and clinical parameters were developed and compared. The combination of radiomics features with clinical parameters achieved the best results, stratifying patients into high- and low-risk groups with significantly different outcomes.
Article
Radiology, Nuclear Medicine & Medical Imaging
Pouria Yazdian Anari et al.
Summary: In this study, a machine learning algorithm was used to predict the growth rate category of renal tumors in VHL patients based on radiomic features. The results provide a personalized approach for surveillance and surgical planning.
ABDOMINAL RADIOLOGY
(2022)
Review
Urology & Nephrology
Matteo Ferro et al.
Summary: This article provides an overview of the application of radiomics in prostate cancer patients, emphasizing its potential for personalized treatment and future applications.
THERAPEUTIC ADVANCES IN UROLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ruben Ngnitewe Massa'a et al.
Summary: Machine learning models trained with MRI-based radiomics features can accurately distinguish benign from malignant solid renal masses. These models have the potential to improve the clinical management of solid renal masses and reduce the frequency of unnecessary biopsies or surgical resections.
ABDOMINAL RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Taek Min Kim et al.
Summary: This study aims to assess the CT findings of renal epithelioid angiomyolipoma (EAML) and develop a radiomics-based model for differentiating EAMLs and clear cell renal cell carcinomas (RCCs). The shape of the tumor and the tumor-to-cortex enhancement ratio were found to be significant factors for identifying EAMLs. The radiomics model showed better diagnostic performance than an inexperienced radiologist.
ABDOMINAL RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Jeroen Bleker et al.
Summary: The deep learning masked auto-fixed volume of interest (DLM) segmentation method provides a more accurate radiomics diagnosis of clinically significant prostate cancer (CS PCa) compared to manual segmentation, while significantly reducing expert time investment.
EUROPEAN RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Shunya Matsumoto et al.
Summary: This study investigates the utility of radiomics features of diffusion-weighted magnetic resonance imaging (DW-MRI) in differentiating fat-poor angiomyolipoma (fpAML) from clear cell renal cell carcinoma (ccRCC). The results suggest that radiomics features of ADC maps are useful for distinguishing fpAML from ccRCC. Certain radiomics features, including mean ADC value and higher-texture features, play an important role in the differentiation. This study provides valuable insights for the diagnosis of renal tumors.
ABDOMINAL RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Xu Bai et al.
Summary: The study evaluated the performance of a multiparametric MRI radiomics-based nomogram for individualized prediction of SDM in ccRCC patients, showing good predictive ability in the training cohort and internal and external validation cohorts, as well as effective prediction in different sizes of ccRCC subgroups.
EUROPEAN RADIOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Li Zhang et al.
Summary: The study explored the value of biparametric MRI-based radiomics signatures for preoperative prediction of prostate cancer grade compared with radiologist visual assessments based on PI-RADS V2.1 scores. Machine learning analysis of MRI radiomics models proved to be helpful in distinguishing high-grade from low-grade prostate cancer, outperforming PI-RADS V2.1 scores. Among the three machine learning algorithms, the random forest tree model showed slightly better diagnostic performance.
CONTRAST MEDIA & MOLECULAR IMAGING
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Zaosong Zheng et al.
Summary: Radiomics analysis was used to develop a CT-based nomogram for preoperative prediction of nuclear grades in ccRCC, showing good accuracy in discriminating nuclear grades and providing individualized treatment for patients.
EUROPEAN RADIOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Renato Cuocolo et al.
Summary: The study aimed to establish a machine learning model using radiomics features extracted from prostate MRI index lesions to detect extraprostatic extension of prostate cancer. The model demonstrated high accuracy in a multicenter setting and could assist radiologists in EPE detection.
EUROPEAN RADIOLOGY
(2021)
Article
Oncology
Valentina Giannini et al.
Summary: The study developed and validated a CAD system for detecting and characterizing PCas based on aggressiveness. Using artificial intelligence algorithms, the system effectively distinguished between low and high aggressive tumors, showing promise as a predictive tool for personalized decision making in patients with PCa.
FRONTIERS IN ONCOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Stefanie J. Hectors et al.
Summary: A machine learning model based on T2WI radiomics features was constructed to identify clinically significant prostate cancer in PI-RADS 3 lesions, showing good performance with statistically significant results.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Sidharth Gurbani et al.
Summary: This study aimed to evaluate the use of CT radiomics features and machine learning analysis to identify aggressive tumor features in large renal cell carcinomas. The results showed weak but statistically significant associations between high nuclear grade and sarcomatoid features in large RCC. The XG Boost Model performed best on the tested data, with different radiomics features and AUC values for NG and sarc models.
ABDOMINAL RADIOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Zongtai Zheng et al.
Summary: The study aimed to construct a radiomics-clinical nomogram for preoperatively differentiating high-grade from low-grade bladder cancer. Results showed that the radiomics-clinical nomogram incorporating radiomics features with VI-RADS score demonstrated good diagnostic power in both the training and validation sets.
ABDOMINAL RADIOLOGY
(2021)
Article
Oncology
Zongtai Zheng et al.
Summary: A nomogram based on mpMRI radiomics signature and VI-RADS score was constructed to differentiate between MIBC and NMIBC before surgery. The nomogram showed good discrimination and calibration, improving clinical decision making in bladder cancer patients. The radiomics signature may be a useful tool for preoperative differentiation of muscle-invasive status in bladder cancer.
FRONTIERS IN ONCOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Moozhan Nikpanah et al.
Summary: The study found that the deep convolutional neural network showed high accuracy and reliability in differentiating ccRCC from oncocytoma. The observer study showed moderate agreement between two radiologists and the AI algorithm.
Article
Radiology, Nuclear Medicine & Medical Imaging
Christopher S. Lim et al.
Summary: Machine learning of radiomic features extracted from T2W and ADC achieved moderate accuracy in determining whether PI-RADS category 3 lesions represent PCa.
ABDOMINAL RADIOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Xiaoli Li et al.
Summary: This study developed a radiomics nomogram for preoperative differentiation of RO and ccRCC based on clinical data and radiomics signature, showing promising diagnostic performance. The nomogram could potentially reduce unnecessary surgeries and assist in tailoring precise therapy.
ABDOMINAL RADIOLOGY
(2021)
Article
Oncology
Gumuyang Zhang et al.
Summary: The DL model based on CT images showed good prediction ability for muscle-invasive status of BCa, outperforming subjective assessments, and may potentially improve individual treatment outcomes.
FRONTIERS IN ONCOLOGY
(2021)
Review
Urology & Nephrology
Veeru Kasivisvanathan et al.
Article
Surgery
Fuad F. Elkhoury et al.
Article
Radiology, Nuclear Medicine & Medical Imaging
Soleen Ghafoor et al.
JOURNAL OF NUCLEAR MEDICINE
(2019)
Article
Oncology
Rebecca L. Siegel et al.
CA-A CANCER JOURNAL FOR CLINICIANS
(2019)
Review
Medicine, General & Internal
Mark S. Litwin et al.
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION
(2017)
Article
Medicine, General & Internal
Hashim U. Ahmed et al.
Review
Medicine, General & Internal
Ashish M. Kamat et al.
Article
Medicine, General & Internal
M. Minhaj Siddiqui et al.
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION
(2015)
Article
Radiology, Nuclear Medicine & Medical Imaging
Berrend G. Muller et al.
Article
Oncology
Philippe Lambin et al.
EUROPEAN JOURNAL OF CANCER
(2012)
Article
Radiology, Nuclear Medicine & Medical Imaging
Nicole Hindman et al.
