Browsing by Author "Nunes, Rita G."
Now showing 1 - 7 of 7
Results Per Page
Sort Options
- Apparent diffusion coefficient in the analysis of prostate cancer: determination of optimal b-value pair to differentiate normal from malignant tissuePublication . Adubeiro, Nuno; Nogueira, Maria Luísa; Nunes, Rita G.; Ferreira, Hugo Alexandre; Ribeiro, Eduardo; La Fuente, José Maria FerreiraPurpose Determining optimal b-value pair for differentiation between normal and prostate cancer (PCa) tissues. Methods Forty-three patients with diagnosis or PCa symptoms were included. Apparent diffusion coefficient (ADC) was estimated using minimum and maximum b-values of 0, 50, 100, 150, 200, 500 s/mm2 and 500, 800, 1100, 1400, 1700 and 2000s/mm2, respectively. Diagnostic performances were evaluated when Area-under-the-curve (AUC) > 95%. Results 15 of the 35 b-values pair surpassed this AUC threshold. The pair (50, 2000 s/mm2) provided the highest AUC (96%) with ADC cutoff 0.89 × 10–3 mm2/s, sensitivity 95.5%, specificity 93.2% and accuracy 94.4%. Conclusions The best b-value pair was b = 50, 2000 s/mm2.
- Breast DWI at 3 T: influence of the fat-suppression technique on image quality and diagnostic performancePublication . Nogueira, Luisa; Brandão, Sofia; Nunes, Rita G.; Ferreira, Hugo Alexandre; Loureiro, Joana; Ramos, IsabelAim To evaluate two fat-suppression techniques: short tau inversion recovery (STIR) and spectral adiabatic inversion recovery (SPAIR) regarding image quality and diagnostic performance in diffusion-weighted imaging (DWI) of breast lesions at 3 T. Materials and methods Ninety-two women (mean age 48 ± 12.1 years; range 21–78 years) underwent breast MRI. Two DWI pulse sequences, with b-values (50 and 1000 s/mm2) were performed with STIR and SPAIR. The signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), suppression homogeneity, and apparent diffusion coefficient (ADC) values were quantitatively assessed for each technique. Values were compared between techniques and lesion type. Receiver operating characteristics (ROC) analysis was used to evaluate lesion discrimination. Results One hundred and fourteen lesions were analysed (40 benign and 74 malignant). SNR and CNR were significantly higher for DWI-SPAIR; fat-suppression uniformity was better for DWI-STIR (p < 1 × 10−4). ADC values for benign and malignant lesions and normal tissue were 1.92 × 10−3, 1.18 × 10−3, 1.86 × 10−3 s/mm2 for DWI-STIR and 1.80 × 10−3, 1.11 × 10−3, 1.79 × 10−3 s/mm2 for SPAIR, respectively. Comparison between fat-suppression techniques showed significant differences in mean ADC values for benign (p = 0.013) and malignant lesions (p = 0.001). DWI-STIR and -SPAIR ADC cut-offs were 1.42 × 10−3 and 1.46 × 10−3 s/mm2, respectively. Diagnostic performance for DWI-STIR versus SPAIR was: accuracy (81.6 versus 83.3%), area under curve (87.7 versus 89.2%), sensitivity (79.7 versus 85.1%), and specificity (85 versus 80%). Positive predictive value was similar. Conclusion The fat-saturation technique used in the present study may influence image quality and ADC quantification. Nevertheless, STIR and SPAIR techniques showed similar diagnostic performances, and therefore, both are suitable for use in clinical practice.
- Diffusion MRI outside the brainPublication . Nunes, Rita G.; Nogueira, Luisa; Gaspar, Andreia S.; Adubeiro, Nuno; Brandão, SofiaThis manuscript provides an overview of recent developments in Diffusion-Weighted Imaging (DWI) outside the brain, focusing on liver, breast, prostate, muskuloskeletal (MSK) and cardiac applications. A general introduction to cross-cutting acquisition and image processing challenges is first provided. These often include short T2 relaxation times, the need to image a large field-of-view with the resulting complications in shimming the B0 field and achieving good fat suppression. Some of the strategies developed for dealing with motion, namely cardiac and respiratory motion are described. Specific sections are then presented for each of the aforementioned organs. A motivation for the clinical applicability of DWI is first provided, followed by specific image acquisition and processing considerations. Quantitative imaging is becoming standard in clinical practice, and the Apparent Diffusion Coefficient is routinely estimated in the liver, breast and prostate. Application of alternative signal models in these organs is being explored, including both the Intravoxel Incoherent Motion and Diffusion Kurtosis models. Ongoing efforts are focused on evaluating the potential clinical added value of the extra parameters and on improving their repeatability. MSK and cardiac DWI have shown potential for assessing pathological changes in fiber architecture, but further validation is required to enable application in the clinical setting.
- Diffusion-Weighted Breast Imaging: Beyond MorphologyPublication . Nogueira, Luísa; Nunes, Rita G.; Brandão, Sofia; Ramos, IsabelDiffusion-weighted imaging (DWI) is a magnetic resonance imaging (MRI) technique that explores the molecular diffusivity of water in biological tissues to probe its microstructure. Its application to the study of breast lesions has been shown to improve their detection, characterization, and the diagnostic accuracy of breast lesions using MRI. In this chapter, the biophysical basis of diffusion is presented, including the model currently used for DWI in the clinical setting; the concept of apparent diffusion coefficient (ADC) is introduced. A theoretical framework of DWI in healthy conditions and in tissues affected by pathological processes is presented, followed by a literature review on the application of DWI to breast imaging. As the technique has only recently been used in breast imaging studies, controversial issues regarding its application have arisen, namely related to its technical challenges. Therefore, we detail the main technical issues associated with the implementation of DWI in the clinical setting and present potential approaches for obtaining good-quality images. Finally, we identify relevant future research needs involving hardware and software optimization as well as clinical issues which need to be addressed to improve breast lesion diagnosis.
- Editorial for “3D breast cancer segmentation in DCE‐MRI using deep learning with weak annotation”Publication . Nogueira, Luísa; Adubeiro, Nuno; Nunes, Rita G.Magnetic resonance imaging (MRI) shows higher diagnostic performance in the detection of breast tumors, compared with other imaging modalities. Breast MRI protocols include dynamic contrast-enhanced (DCE) images with high spatial and temporal resolution and are central indiagnosis, staging, and follow-up of breast cancer. DCE features provide physiological and anatomical lesion characteristics. To extract these data, manual lesion segmentation is currently performed, which is a critical time-consuming step,introducing bias and variability and impacting the reproducibility of the extracted features. To overcome these limitations, artificial intelligence algorithms have been explored, especially deep learning (DL) methods, for automatic lesion segmentation. This has been an active area of research, pivotal in the analysis of quantitative medical images. Most lesion segmentation methods have been based on semiautomatic or supervised learning approaches, presenting an important limitation: slice-by-slice 2D segmentations are typically performed, leading to suboptimal 3D masks upon concatenation. Recently, DL methods based on vision transformers have gained popularity in breast lesion segmentation, improving results over traditional machine learning. Although fully convolutional neural networks (CNNs) show powerful learning capabilities, their performance in learning long-range dependencies is limited, presenting decreased capacity in the segmentation of structures including different shapes and scales. UNETR is an architecture that replaces the CNN-based encoder with a transformer, which can capture low-level details in 3D segmentation. UNETR directly connects the encoder to the decoder via skip connections and can directly use volumetric data. Compared with CNN or transformer-based segmentation methods, UNETR can better capture dependencies at diverse spatial scales, both local and long-range enabling improved segmentation. In this retrospective study, Kim et al developed a model based on weak annotations, for detection and 3D segmentation of breast cancer in a sample of 736 women, using different input combinations in a three-time point (3TP) approach, from DCE-MRI images, acquired in two 3 T scanners from different manufacturers. The sample was divided into training (N = 544)and test sets (N = 192). To reduce the workload required toobta in ground truth segmentations, tumors were first segmented using weak annotations by two radiologists in consensus drawing bounding rectangles encompassing the lesion on two projection images. The rectangles were used to generate a 3D bounding box applied to the image obtained by subtracting the pre-contrast from the post-contrast image. An automatic thres holding method was used for automatic lesion segmentation; the mask was then refined to better define the lesion boundaries and exclude noisyor confounding regions (false positives). For training the segmen-tation network, images acquired at three different temporal acquisition points (pre-contrast, early, and delayed post-contrast) were used to construct three inputs: input 1 (pre-contrast, early phase),input 2 (pre-contrast, early, and delayed phase), and input 3 (pre-contrast and delayed phase). A different UNETR model wastrained for each input, and segmentation performances were compared, qualitatively and quantitatively, based on MRI features and immunohistochemical (IHC) classification. The best DL model presented a reliable performance for automated 3D segmentation of breast cancer with a median dicesimilarity coefficient (DSC) of 0.75 for the whole breast and 0.89 for the index lesion. The performance of the UNETR model was in accordance with the DSC values reported by other researchers employing alternative segmentation algorithms. Regarding the qualitative analysis of the segmentation results, the segmentation was successfully done in 83% of the cases derived from inputs 1 and 2, and from these, 95% were considered as acceptable detection. The authors also evaluated the performance of the segmentation according to base line characteristics and found significant differences for the whole breast and main lesion. For main lesion, significant differences were observed according to lesion size and IHC type. Regarding visual analysis, significant differences were found between lesion type (mass vs. non-mass enhancement) and background parenchymal enhancement (BPE) level. In their study, there were nine cases of failed segmentation, which corresponded totumors with small volumes, from which five cases were not segmented and four cases corresponded to abundant BPE,meaning false-positive results.© 2023 International Society for Magnetic Resonance in Medicine. 2263 Further developments of 3D UNETR architecture could be done to improve small lesion detection, to distinguish between mass and non-mass lesions, especially the boundaries of non-masses, and to distinguish between BPE and tumors. Attending to the implementation of DL algorithms in the clinical practice, this type of algorithm is expected to improve the detection of small lesions and the prediction of response to treatment, there by reducing the number of performed biopsies and, potentially, enabling the use of an abbreviated MRI pro-tocol, which would reduce MRI exam durations, improving patient comfort, and reducing costs.
- Editorial for “Three‐dimensional multifrequency MR elastography for microvascular invasion and prognosis assessment in Hepatocellular Carcinoma”Publication . Adubeiro, Nuno; Nunes, Rita G.; Nogueira, LuísaThe prognosis of individuals with hepatocellular carcinoma(HCC), the most prevalent primary liver malignancy, is closely linked to the aggressiveness and recurrence of the tumor. The occurrence of complications after surgery continues to be a major and persistent challenge. MR elastography (MRE) employs a modified phase-contrast imaging sequence, combined with the use of an external driver to transmit mechanical vibrations to the tissues, to identify propagating shear waves within the liver. This technique allows the assessment of a substantial portion of the liver and provides information on multiple mechanical properties associated with various pathophysiological states. Due to substantial progress in MR technology, MRE has proven to be a precise noninvasive diagnostic method for detecting and monitoring various liver diseases. MRE imaging could serve as a valuable tool for staging malignancy and predict disease prognosis.
- Gamma distribution model in breast cancer diffusion-weighted imagingPublication . Borlinhas, Filipa; Nogueira, Luisa; Brandao, Sofia; Nunes, Rita G.; Loureiro, Joana; Ramos, Isabel; Ferreira, Hugo ASummary form only given. Many diffusion models have been proposed in order to obtain more information from breast tumor tissues through Magnetic Resonance Imaging (MRI) (1). The Gamma distribution (GD) may model MRI signal decay based on a statistical approach. This model considers the Theta parameter, which indicates the statistical dispersion of the distribution, and the k parameter, which is responsible for the probability distribution shape. If Theta shows higher values, then there will be a more spread out distribution and if k shows lower values the distribution shape will be more affected, which would be expected in malignant tumors due to tissue heterogeneity (1). The purpose of this study was to evaluate if GD model is capable of distinguishing between different breast tumors. Materials and Methods: In this study 85 breast tumor lesions were analyzed, including 17 benign lesions (Fibroadenoma, FA) and 68 malignant lesions (43 Invasive Ductal Carcinomas, IDC 19 Invasive Lobular Carcinomas, ILC and 6 Ductal Carcinoma in situ, CDIS). Informed consent was obtained for all patients. Data were acquired using a 3T MRI scanner with a dedicated breast coil and a DWI sequence with 3 orthogonal diffusion gradient directions and 8 b values between 0 and 3000s/mm 2 . Theta and k parameters were acquired from fitting data to the GD model, and mean values were obtained to compare between benign and malignant lesions, and between histological types. Non-parametric statistics were used (α=0.05). Results and Discussion: Significantly lower Theta and higher k values were observed in benign lesions ((0.65±0.43)×10 -3 mm 2 /s, 4.29±1.90, respectively) when compared to malignant lesions ((0.97±0.50)×10 -3 mm 2 /s, 1.23±0.52, respectively). It was also possible to differentiate FA from IDC lesions with both Theta and k probably due to IDC heterogeneity, which restricts diffusion. Unlike other diffusion model parameters, these were able to differentiate FA and ILC, and FA and CDIS lesions, suggesting that the GD model could bring advantages over other diffusion models in characterizing breast tumors.