Inter-Reader Agreement of Diffusion-Weighted Magnetic Resonance Imaging for Breast Cancer Detection: A Multi-Reader Retrospective Study

Pesapane, Filippo; Rotili, Anna; Penco, Silvia; Montesano, Marta; Agazzi, Giorgio Maria; Dominelli, Valeria; Trentin, Chiara; Pizzamiglio, Maria; Cassano, Enrico

doi:10.3390/cancers13081978

Cited by 10 publications

(15 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image processing in the medical field is also used with great results for other types of image. For example in [ 25 , 26 ] a diffusion-weighted magnetic resonance imaging is discussed for breast and lung cancer detection. Reference [ 27 ] presented a survey of deep learning algorithms used in medical applications.…”

Section: Related Workmentioning

confidence: 99%

Automatic Detection of Colorectal Polyps Using Transfer Learning

Dulf

Bledea

Mocan

et al. 2021

Sensors

View full text Add to dashboard Cite

Colorectal cancer is the second leading cause of cancer death and ranks third worldwide in diagnosed malignant pathologies (1.36 million new cases annually). An increase in the diversity of treatment options as well as a rising population require novel diagnostic tools. Current diagnostics involve critical human thinking, but the decisional process loses accuracy due to the increased number of modulatory factors involved. The proposed computer-aided diagnosis system analyses each colonoscopy and provides predictions that will help the clinician to make the right decisions. Artificial intelligence is included in the system both offline and online image processing tools. Aiming to improve the diagnostic process of colon cancer patients, an application was built that allows the easiest and most intuitive interaction between medical staff and the proposed diagnosis system. The developed tool uses two networks. The first, a convolutional neural network, is capable of classifying eight classes of tissue with a sensitivity of 98.13% and an F1 score of 98.14%, while the second network, based on semantic segmentation, can identify the malignant areas with a Jaccard index of 75.18%. The results could have a direct impact on personalised medicine combining clinical knowledge with the computing power of intelligent algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Automatic Detection of Colorectal Polyps Using Transfer Learning

Dulf

Bledea

Mocan

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In contrast to fatty breasts, dense breasts did not affect lesion detectability and ADC values [ 75 ]. It was found that for DWI, expert-level readers are necessary for reaching good results [ 76 ], because the detection and avoidance of areas affected by artefacts is important for parameter analysis and requires some experience. In general, DWI is not suited for patients with implants and in case of small lesions, which is a potential source of bias.…”

Section: Discussionmentioning

confidence: 99%

Simplified intravoxel incoherent motion DWI for differentiating malignant from benign breast lesions

et al. 2022

View full text Add to dashboard Cite

Background To evaluate simplified intravoxel incoherent motion (IVIM) diffusion-weighted imaging (DWI) for differentiating malignant versus benign breast lesions as (i) stand-alone tool and (ii) add-on to dynamic contrast-enhanced magnetic resonance imaging. Methods 1.5-T DWI data (b = 0, 50, 250, 800 s/mm2) were retrospectively analysed for 126 patients with malignant or benign breast lesions. Apparent diffusion coefficient (ADC) ADC (0, 800) and IVIM-based parameters D1′ = ADC (50, 800), D2′ = ADC (250, 800), f1′ = f (0, 50, 800), f2′ = f (0, 250, 800) and D*′ = D* (0, 50, 250, 800) were voxel-wise calculated without fitting procedures. Regions of interest were analysed in vital tumour and perfusion hot spots. Beside the single parameters, the combined use of D1′ with f1′ and D2′ with f2′ was evaluated. Lesion differentiation was investigated for lesions (i) with hyperintensity on DWI with b = 800 s/mm2 (n = 191) and (ii) with suspicious contrast-enhancement (n = 135). Results All lesions with suspicious contrast-enhancement appeared also hyperintense on DWI with b = 800 s/mm2. For task (i), best discrimination was reached for the combination of D1′ and f1′ using perfusion hot spot regions-of-interest (accuracy 93.7%), which was higher than that of ADC (86.9%, p = 0.003) and single IVIM parameters D1′ (88.0%) and f1′ (87.4%). For task (ii), best discrimination was reached for single parameter D1′ using perfusion hot spot regions-of-interest (92.6%), which were slightly but not significantly better than that of ADC (91.1%) and D2′ (88.1%). Adding f1′ to D1′ did not improve discrimination. Conclusions IVIM analysis yielded a higher accuracy than ADC. If stand-alone DWI is used, perfusion analysis is of special relevance.

show abstract

“…Image acquisition details of our 1.5 T MRI scanner protocol are reported in the Table S1 [ 19 , 20 ]. The dynamic study consisted of three-dimensional T1-w gradient-echo sequences acquired once before and five times after intravenous administration of 0.1 mmol/kg of a gadolinium chelate at 90 s temporal resolution.…”

Section: Methodsmentioning

confidence: 99%

Radiomics of MRI for the Prediction of the Pathological Response to Neoadjuvant Chemotherapy in Breast Cancer Patients: A Single Referral Centre Analysis

Pesapane

Rotili

Botta

et al. 2021

Cancers

Self Cite

View full text Add to dashboard Cite

Objectives: We aimed to determine whether radiomic features extracted from a highly homogeneous database of breast MRI could non-invasively predict pathological complete responses (pCR) to neoadjuvant chemotherapy (NACT) in patients with breast cancer. Methods: One hundred patients with breast cancer receiving NACT in a single center (01/2017–06/2019) and undergoing breast MRI were retrospectively evaluated. For each patient, radiomic features were extracted within the biopsy-proven tumor on T1-weighted (T1-w) contrast-enhanced MRI performed before NACT. The pCR to NACT was determined based on the final surgical specimen. The association of clinical/biological and radiomic features with response to NACT was evaluated by univariate and multivariable analysis by using random forest and logistic regression. The performances of all models were assessed using the areas under the receiver operating characteristic curves (AUC) with 95% confidence intervals (CI). Results: Eighty-three patients (mean (SD) age, 47.26 (8.6) years) were included. Patients with HER2+, basal-like molecular subtypes and Ki67 ≥ 20% presented a pCR to NACT more frequently; the clinical/biological model’s AUC (95% CI) was 0.81 (0.71–0.90). Using 136 representative radiomics features selected through cluster analysis from the 1037 extracted features, a radiomic score was calculated to predict the response to NACT, with AUC (95% CI): 0.64 (0.51–0.75). After combining the clinical/biological and radiomics models, the AUC (95% CI) was 0.83 (0.73–0.92). Conclusions: MRI-based radiomic features slightly improved the pre-treatment prediction of pCR to NACT, in addiction to biological characteristics. If confirmed on larger cohorts, it could be helpful to identify such patients, to avoid unnecessary treatment.

show abstract

Inter-Reader Agreement of Diffusion-Weighted Magnetic Resonance Imaging for Breast Cancer Detection: A Multi-Reader Retrospective Study

Cited by 10 publications

References 35 publications

Automatic Detection of Colorectal Polyps Using Transfer Learning

Automatic Detection of Colorectal Polyps Using Transfer Learning

Simplified intravoxel incoherent motion DWI for differentiating malignant from benign breast lesions

Radiomics of MRI for the Prediction of the Pathological Response to Neoadjuvant Chemotherapy in Breast Cancer Patients: A Single Referral Centre Analysis

Contact Info

Product

Resources

About