Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images

Frontiers in Endocrinology, Journal Year: 2024, Volume and Issue: 15

Published: March 28, 2024

Purpose To develop and validate a deep learning radiomics (DLR) model that uses X-ray images to predict the classification of osteoporotic vertebral fractures (OVFs). Material methods The study encompassed cohort 942 patients, involving examinations 1076 vertebrae through X-ray, CT, MRI across three distinct hospitals. OVFs were categorized as class 0, 1, or 2 based on Assessment System Thoracolumbar Osteoporotic Fracture. dataset was divided randomly into four subsets: training set comprising 712 samples, an internal validation with 178 external containing 111 prospective consisting 75 samples. ResNet-50 architectural used implement transfer (DTL), undergoing -pre-training separately RadImageNet ImageNet datasets. Features from DTL extracted integrated using images. optimal fusion feature identified least absolute shrinkage selection operator logistic regression. Evaluation predictive capabilities for involved eight machine models, assessed receiver operating characteristic curves employing “One-vs-Rest” strategy. Delong test applied compare performance superior against model. Results Following pre-training datasets, yielded 17 12 features, respectively. Logistic regression emerged algorithm both DLR models. Across set, macro-average Area Under Curve (AUC) surpassed those dataset, statistically significant differences observed (P<0.05). Utilizing binary strategy, demonstrated efficacy in predicting Class achieving AUC 0.969 accuracy 0.863. Predicting 1 0.945 0.875, while 2, 0.809 0.692, Conclusion model, outperformed OVFs, generalizability confirmed set.

Language: Английский

---

Artificial intelligence in breast cancer imaging: risk stratification, lesion detection and classification, treatment planning and prognosis—a narrative review

Exploration of Targeted Anti-tumor Therapy, Journal Year: 2022, Volume and Issue: unknown, P. 795 - 816

Published: Dec. 27, 2022

The advent of artificial intelligence (AI) represents a real game changer in today's landscape breast cancer imaging. Several innovative AI-based tools have been developed and validated recent years that promise to accelerate the goal patient-tailored management. Numerous studies confirm proper integration AI into existing clinical workflows could bring significant benefits women, radiologists, healthcare systems. approach has proved particularly useful for developing new risk prediction models integrate multi-data streams planning individualized screening protocols. Furthermore, help radiologists pre-screening lesion detection phase, increasing diagnostic accuracy, while reducing workload complications related overdiagnosis. Radiomics radiogenomics approaches extrapolate so-called imaging signature tumor plan targeted treatment. main challenges development are huge amounts high-quality data required train validate these need multidisciplinary team with solid machine-learning skills. purpose this article is present summary most important applications imaging, analyzing possible perspectives widespread adoption tools.

Language: Английский

---

Machine Learning Approaches with Textural Features to Calculate Breast Density on Mammography

Current Oncology, Journal Year: 2023, Volume and Issue: 30(1), P. 839 - 853

Published: Jan. 7, 2023

breast cancer (BC) is the world's most prevalent in female population, with 2.3 million new cases diagnosed worldwide 2020. The great efforts made to set screening campaigns, early detection programs, and increasingly targeted treatments led significant improvement patients' survival. Full-Field Digital Mammograph (FFDM) considered gold standard method for diagnosis of BC. From several previous studies, it has emerged that density (BD) a risk factor development BC, affecting periodicity plans present today at an international level.in this study, focus mammographic image processing techniques allow extraction indicators derived from textural patterns mammary parenchyma indicative BD factors.a total 168 patients were enrolled internal training test while 51 compose external validation cohort. Different Machine Learning (ML) have been employed classify breasts based on values tissue density. Textural features extracted only which train classifiers, thanks aid ML algorithms.the accuracy different tested classifiers varied between 74.15% 93.55%. best results reached by Support Vector (accuracy 93.55% percentage true positives negatives equal TPP = 94.44% TNP 92.31%). was not influenced choice selection approach. Considering cohort, SVM, as classifier 7 selected wrapper method, showed 0.95, sensitivity 0.96, specificity 0.90.our preliminary Radiomics analysis approach us objectively identify BD.

Language: Английский

---

Barriers and facilitators of artificial intelligence conception and implementation for breast imaging diagnosis in clinical practice: a scoping review

European Radiology, Journal Year: 2023, Volume and Issue: 34(3), P. 2096 - 2109

Published: Sept. 2, 2023

Although artificial intelligence (AI) has demonstrated promise in enhancing breast cancer diagnosis, the implementation of AI algorithms clinical practice encounters various barriers. This scoping review aims to identify these barriers and facilitators highlight key considerations for developing implementing solutions imaging.

Language: Английский

---

Risk Assessment and Pancreatic Cancer: Diagnostic Management and Artificial Intelligence

Cancers, Journal Year: 2023, Volume and Issue: 15(2), P. 351 - 351

Published: Jan. 5, 2023

Pancreatic cancer (PC) is one of the deadliest cancers, and it responsible for a number deaths almost equal to its incidence. The high mortality rate correlated with several explanations; main late disease stage at which majority patients are diagnosed. Since surgical resection has been recognised as only curative treatment, PC diagnosis initial believed tool improve survival. Therefore, patient stratification according familial genetic risk creation screening protocol by using minimally invasive diagnostic tools would be appropriate. cystic neoplasms (PCNs) subsets lesions deserve special management avoid overtreatment. current programs based on annual employment magnetic resonance imaging cholangiopancreatography sequences (MR/MRCP) and/or endoscopic ultrasonography (EUS). For unfit MRI, computed tomography (CT) could proposed, although CT results in lower detection rates, compared small lesions. actual major limit incapacity detect characterize pancreatic intraepithelial neoplasia (PanIN) EUS MR/MRCP. possibility utilizing artificial intelligence models evaluate higher-risk favour these entities, more data needed support real utility applications field screening. motives, appropriate realize research settings.

Language: Английский

---

Multiparametric MRI-Based Interpretable Radiomics Machine Learning Model Differentiates Medulloblastoma and Ependymoma in Children: A Two-Center Study

Academic Radiology, Journal Year: 2024, Volume and Issue: 31(8), P. 3384 - 3396

Published: March 20, 2024

Language: Английский

---

Ability of Delta Radiomics to Predict a Complete Pathological Response in Patients with Loco-Regional Rectal Cancer Addressed to Neoadjuvant Chemo-Radiation and Surgery

Cancers, Journal Year: 2022, Volume and Issue: 14(12), P. 3004 - 3004

Published: June 18, 2022

We performed a pilot study to evaluate the use of MRI delta texture analysis (D-TA) as methodological item able predict frequency complete pathological responses and, consequently, outcome patients with locally advanced rectal cancer addressed neoadjuvant chemoradiotherapy (C-RT) and subsequently, radical surgery. In particular, we carried out retrospective including 100 adenocarcinoma who received C-RT then surgery in three different oncological institutions between January 2013 December 2019. Our experimental design was focused on evaluation gross tumor volume (GTV) at baseline after by means MRI, which contoured T2, DWI, ADC sequences. Multiple parameters were extracted using LifeX Software, while D-TA calculated percentage variations two time points. Both univariate multivariate (logistic regression) were, therefore, order correlate above-mentioned TA examined patients’ population focusing detection response (pCR, no viable cells: TRG 1) main statistical endpoint. ROC curves datasets considering that 21 patients, only 21% achieved an actual pCR. our training dataset series, pCR significantly correlated GLCM-Entropy only, when binary logistic (AUC for 0.87). A confirmative regression repeated remaining validation 0.92 0.88, respectively). Overall, these results support hypothesis may have significant predictive value detecting occurrence patient series. If confirmed prospective multicenter trials, critical role selection benefit form chemoradiotherapy.

Language: Английский

---

Radiomics in Lung Metastases: A Systematic Review

Journal of Personalized Medicine, Journal Year: 2023, Volume and Issue: 13(2), P. 225 - 225

Published: Jan. 27, 2023

Due to the rich vascularization and lymphatic drainage of pulmonary tissue, lung metastases (LM) are not uncommon in patients with cancer. Radiomics is an active research field aimed at extraction quantitative data from diagnostic images, which can serve as useful imaging biomarkers for a more effective, personalized patient care. Our purpose illustrate current applications, strengths weaknesses radiomics lesion characterization, treatment planning prognostic assessment LM, based on systematic review literature.

Language: Английский

---

CT-Based Radiomics Predicts the Malignancy of Pulmonary Nodules: A Systematic Review and Meta-Analysis

Academic Radiology, Journal Year: 2023, Volume and Issue: 30(12), P. 3064 - 3075

Published: June 27, 2023

Language: Английский

---

Predicting Axillary Lymph Node Metastasis in Young Onset Breast Cancer: A Clinical-Radiomics Nomogram Based on DCE-MRI

Breast Cancer Targets and Therapy, Journal Year: 2025, Volume and Issue: Volume 17, P. 103 - 113

Published: Jan. 1, 2025

Young onset breast cancer, diagnosed in women under 50, is known for its aggressive nature and challenging prognosis. Precisely forecasting axillary lymph node metastasis (ALNM) essential customizing treatment plans enhancing patient results. This research sought to create verify a clinical-radiomics nomogram that combines radiomic features from Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) with standard clinical predictors improve the accuracy of predicting ALNM young cancer patients. We performed retrospective analysis at one facility, involving creation validation two stages.At first, medical model was developed utilizing conventional indicators like tumor dimensions, molecular classifications, multifocal presence, MRI-determined ALN status.A more detailed subsequently by integrating characteristics derived DCE-MRI images.These models were created using logistic regression analyses on training dataset, their effectiveness assessed measuring area receiver operating characteristic curve (AUC) separate dataset. The surpassed clinical-only model, recording an AUC 0.892 dataset 0.877 dataset.Significant included MRI-reported status select features, which markedly enhanced model's predictive capacity. Integrating significantly improves prediction providing valuable tool personalized planning. study underscores potential merging advanced imaging data insights refine oncological models. Future should expand multicentric studies include genomic boost nomogram's generalizability precision.

Language: Английский

---