Apparent Diffusion Coefficient Maps Research Articles

Radiomic Features Are Predictive of Response in Rectal Cancer Undergoing Therapy.

Rectal cancer is a major mortality cause in the United States (US), and its treatment is based on individual risk factors for recurrence in each patient. In patients with rectal cancer, accurate assessment of response to chemoradiotherapy has increased in importance as the variety of treatment options has grown. In this scenario, a controversial non-operative approach may be considered in some patients for whom complete tumor regression is believed to have occurred. The recommended treatment for locally advanced rectal cancer (LARC, T3-4 ± N+) is total mesorectal excision (TME) after neoadjuvant chemoradiotherapy (nCRT). Magnetic resonance imaging (MRI) has become a standard technique for local staging of rectal cancer (tumor, lymph node, and circumferential resection margin [CRM] staging), in both the US and Europe, and it is getting widely used for restaging purposes. In our study, we aimed to use an MRI radiomic model to identify features linked to the different responses of chemoradiotherapy of rectal cancer before surgery, and whether these features are helpful to understand the effectiveness of the treatments. We retrospectively evaluated adult patients diagnosed with LARC who were subjected to at least 2 MRI examinations in 10-12 weeks at our hospital, before and after nCRT. The MRI acquisition protocol for the 2 exams included T2 sequence and apparent diffusion coefficient (ADC) map. The patients were divided into 2 groups according to the treatment response: complete or good responders (Group 1) and incomplete or poor responders (Group 2). MRI images were segmented, and quantitative features were extracted and compared between the two groups. Features that showed significant differences (SF) were then included in a LASSO regression method to build a radiomic-based predictive model. We included 38 patients (26 males and 12 females), who are classified from T2 and T4 stages in the rectal cancer TNM. After the nCRT, the patients were divided into Group 1 (13 patients), complete or good responders, and Group 2 (25 patients), incomplete or poor responders. Analysis at baseline generated the following significant features for the Mann-Whitney test (out of a total of 107) for each sequence. Also, the analysis at the end of the follow-up yielded a high number of significant features for the Mann-Whitney test (out of a total of 107) for each image. Features selected by the LASSO regression method for each image analyzed; ROC curves relative to each model are represented. We developed an MRI-based radiomic model that is able to differentiate and predict between responders and non-responders who went through nCRT for rectal cancer. This approach might identify early lesions with high surgical potential from lesions potentially resolving after medical treatment.

Deep Learning Algorithm for Tumor Segmentation and Discrimination of Clinically Significant Cancer in Patients with Prostate Cancer.

We investigated the feasibility of a deep learning algorithm (DLA) based on apparent diffusion coefficient (ADC) maps for the segmentation and discrimination of clinically significant cancer (CSC, Gleason score ≥ 7) from non-CSC in patients with prostate cancer (PCa). Data from a total of 149 consecutive patients who had undergone 3T-MRI and been pathologically diagnosed with PCa were initially collected. The labelled data (148 images for GS6, 580 images for GS7) were applied for tumor segmentation using a convolutional neural network (CNN). For classification, 93 images for GS6 and 372 images for GS7 were used. For external validation, 22 consecutive patients from five different institutions (25 images for GS6, 70 images for GS7) representing different MR machines were recruited. Regarding segmentation and classification, U-Net and DenseNet were used, respectively. The tumor Dice scores for internal and external validation were 0.822 and 0.7776, respectively. As for classification, the accuracies of internal and external validation were 73 and 75%, respectively. For external validation, diagnostic predictive values for CSC (sensitivity, specificity, positive predictive value and negative predictive value) were 84, 48, 82 and 52%, respectively. Tumor segmentation and discrimination of CSC from non-CSC is feasible using a DLA developed based on ADC maps (b2000) alone.

Differentiation of bladder cancer stages using the vesical imaging -reporting and data system and apparent diffusion coefficient.

T stage is closely related to the treatment and prognosis of patients with bladder cancer (BC). However, preoperative T staging is still challenging. Multiparametric magnetic resonance imaging (mpMRI) may be valuable. This study was performed to explore the value of the Vesical Imaging-Reporting and Data System (VI-RADS) and the volumetric apparent diffusion coefficient (ADC) histogram parameters in detecting T2 stage and below stage (≤T2 stage) from T3 stage and above stage (≥T3 stage) BCs. The study included 62 patients (mean age, males vs. females: 62.1±10.9 vs. 61.8±11.7 years) with BC pathologically confirmed by partial or radical cystectomy. All of the tumors were scored normatively by two radiologists using the VI-RADS scoring system by two radiologists. The volumetric ADC histogram of each lesion was obtained from the ADC maps. The Cochran-Armitage test was used to examine the relevance between VI-RADS scores and T stages. The Mann-Whitney U test was used to compare the histogram parameters between ≤T2 stage and ≥T3 stage BCs. A receiver operating characteristic (ROC) curve was used to assess the predictive power of each model. The minimum ADC; mean ADC; median ADC; maximum ADC; and 10th, 25th, 75th, and 90th percentile ADC of ≤T2 stage BCs were significantly higher than those of ≥T3 stage BCs, while skewness and kurtosis had opposite results. VI-RADS achieved the highest area under the curve (AUC) of 0.834 among all parameters. The combination of VI-RADS, skewness and kurtosis yield a significantly higher AUC than VI-RADS alone (0.915 vs. 0.834, P=0.0478). VI-RADS and volume ADC histogram analysis can effectively discriminate between ≤T2 stage and ≥T3 stage BCs, and the volumetric ADC histogram can provide further information to supplement VI-RADS.

Value of CT and MR Imaging in Assessment of Loco-regional Ablated Hepatocellular Carcinoma using LI-RADS Treatment Response Algorithm (version 2018)

Aim: The aim of this work is to evaluate the role of multiphasic CT and dynamic contrast enhanced MRI using LI-RADS treatment response algorithm (version 2018) and the added values of diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) maps in the assessment of loco-Regional ablated Hepatocellular Carcinoma (HCC). Methods: This study included 45 patients with 50 loco-Regional ablated HCC lesions. All patients underwent multiphasic CT scan, diffusion weighted and dynamic contrast enhanced MRI. The reference standards for assessment were based on serum alpha-fetoprotein level and dynamic contrast enhanced MRI. Results: Multi-phasic CT had moderate sensitivity (87.5%) and high specificity (100%) for assessment of HCC LI-RADS treatment response algorithm. Dynamic contrast enhanced MRI had high sensitivity (100%) and specificity (100%) for the assessment of HCC LI-RADS treatment response algorithm. The sensitivity and specificity of DWI to diagnose HCC malignant residual or recurrence were 93.7% and 100%, respectively, with very good performance. The mean ADC values of the malignant residual or recurrence were significantly lower than that of benign post-ablation tissue changes areas and the best ADC cutoff value for differentiation between viable and nonviable lesions was 1.1x10-3 mm²/s. Conclusion: Performance of dynamic contrast enhanced MRI is better than that of multiphasic CT in the assessment of LIRADS treatment response algorithm. DWI and ADC maps could be used as ancillary methods for differentiation between viable and non-viable loco-regional ablated HCC lesions and should be included in LR-TR treatment response algorithm.

Automated acute ischemic stroke lesion delineation based on apparent diffusion coefficient thresholds.

Automated lesion segmentation is increasingly used in acute ischemic stroke magnetic resonance imaging (MRI). We explored in detail the performance of apparent diffusion coefficient (ADC) thresholding for delineating baseline diffusion-weighted imaging (DWI) lesions. Retrospective, exploratory analysis of the prospective observational single-center 1000Plus study from September 2008 to June 2013 (clinicaltrials.org; NCT00715533). We built a fully automated lesion segmentation algorithm using a fixed ADC threshold (≤620 × 10-6 mm2/s) to delineate the baseline DWI lesion and analyzed its performance compared to manual assessments. Diagnostic capabilities of best possible ADC thresholds were investigated using receiver operating characteristic curves. Influential patient factors on ADC thresholding techniques' performance were studied by conducting multiple linear regression. 108 acute ischemic stroke patients were selected for analysis. The median Dice coefficient for the algorithm was 0.43 (IQR 0.20-0.64). Mean ADC values in the DWI lesion (β = -0.68, p < 0.001) and DWI lesion volumes (β = 0.29, p < 0.001) predicted performance. Optimal individual ADC thresholds differed between subjects with a median of ≤691 × 10-6 mm2/s (IQR ≤660-750 × 10-6 mm2/s). Mean ADC values in the DWI lesion (β = -0.96, p < 0.001) and mean ADC values in the brain parenchyma (β = 0.24, p < 0.001) were associated with the performance of individual thresholds. The performance of ADC thresholds for delineating acute stroke lesions varies substantially between patients. It is influenced by factors such as lesion size as well as lesion and parenchymal ADC values. Considering the inherent noisiness of ADC maps, ADC threshold-based automated delineation of very small lesions is not reliable.

Diffusion weighted imaging of extremity bone tumors-inter-reader analysis and incremental value over conventional MR imaging.

To determine whether the addition of diffusion-weighted imaging (DWI) to conventional MRI improves diagnostic accuracy of bone tumor characterization with the hypothesis that the DWI has incremental value in the diagnosis of osseous tumors. In this multireader cross-sectional validation study, four musculoskeletal radiologists evaluated osseous tumors blinded to final diagnosis in two rounds-first without DWI or apparent diffusion coefficient (ADC) maps, then months later with these available. Each reader recorded a binary result as to whether the lesion is benign or malignant. Intraclass correlation (ICC) and Conger's κ were used. Diagnostic performance measures including area under the receiver operating curve (AUC) were reported. 133 osseous tumors of the extremities (76 benign, 57 malignant) were tested. Blinded to DWI, average reader sensitivity, specificity, positive-predictive value, and negative-predictive value were 0.83, 0.92, 0.94, and 0.82, respectively. With DWI, the values were 0.85, 0.92, 0.94, and 0.83, respectively. Interreader agreement was good for both rounds (0.67 and 0.71, respectively, p-value > 0.05). Average reader confidence was 4.1 and 4.4, respectively (p-value < 0.001). ADC values and DWI/ADC ratios showed significant differences between benign and malignant tumors. DWI and ADC show statistically significantly different values of benign from malignant osseous tumors and mildly increased radiologist confidence with similar interreader reliability. However, given similar diagnostic accuracy, conventional MR imaging is adequate for bone tumor characterization and incremental value of DWI is limited. This paper is the first of its kind to report the use of DWI/ADC ratio for the diagnosis of bone tumors.

Shortening Acquisition Time and Improving Image Quality for Pelvic MRI Using Deep Learning Reconstruction for Diffusion-Weighted Imaging at 1.5 T

To determine the impact on acquisition time reduction and image quality of a deep learning (DL) reconstruction for accelerated diffusion-weighted imaging (DWI) of the pelvis at 1.5 Tcompared to standard DWI. A total of 55 patients (mean age, 61±13 years; range, 27-89; 20 men, 35 women) were consecutively included in this retrospective, monocentric study between February and November 2022. Inclusion criteria were (1) standard DWI (DWIS) in clinically indicated magnetic resonance imaging (MRI) at 1.5 Tand (2) DL-reconstructed DWI (DWIDL). All patients were examined using the institution's standard MRI protocol according to their diagnosis including DWI with two different b-values (0and 800s/mm2) and calculation of apparent diffusion coefficient (ADC) maps. Image quality was qualitatively assessed by four radiologists using a visual 5-point Likert scale (5= best) for the following criteria: overall image quality, noise level, extent of artifacts, sharpness, and diagnostic confidence. The qualitative scores for DWIS and DWIDL were compared with the Wilcoxon signed-rank test. The overall image quality was evaluated to be significantly superior in DWIDL compared to DWIS for b =0s/mm2, b =800s/mm2, and ADC maps by all readers (P<.05). The extent of noise was evaluated to be significantly less in DWIDL compared to DWIS for b =0s/mm2, b=800s/mm2, and ADC maps by all readers (P<.001). No significant differences were found regarding artifacts, lesion detectability, sharpness of organs, and diagnostic confidence (P>.05). Acquisition time for DWIS was 2:06minutes, and simulated acquisition time for DWIDL was 1:12minutes. DL image reconstruction improves image quality, and simulation results suggest that a reduction in acquisition time for diffusion-weighted MRI of the pelvis at 1.5 Tis possible.

Problem Solving MRI to Reduce False-Positive Biopsy Related to Breast US: Conductivity vs. DWI vs. Abbreviated Contrast-Enhanced MRI.

While breast ultrasound (US) is a useful tool for diagnosing breast masses, it can entail false-positive biopsy results because of some overlapping features between benign and malignant breast masses and subjective interpretation. To evaluate the performance of conductivity imaging for reducing false-positive biopsy results related to breast US, as compared to diffusion-weighted imaging (DWI) and abbreviated MRI consisting of one pre- and one post-contrast T1-weighted imaging. Prospective. Seventy-nine women (median age, 44 years) with 86 Breast Imaging Reporting and Data System (BI-RADS) category 4 masses as detected by breast US. 3-T, T2-weighted turbo spin echo sequence, DWI, and abbreviated contrast-enhanced MRI (T1-weighted gradient echo sequence). US-guided biopsy (reference standard) was obtained on the same day as MRI. The maximum and mean conductivity parameters from whole and single regions of interest (ROIs) were measured. Apparent diffusion coefficient (ADC) values were obtained from an area with the lowest signal within a lesion on the ADC map. The performance of conductivity, ADC, and abbreviated MRI for reducing false-positive biopsies was evaluated using the following criteria: lowest conductivity and highest ADC values among malignant breast lesions and BI-RADS categories 2 or 3 on abbreviated MRI. One conductivity parameter with the maximum area under the curve (AUC) from receiver operating characteristics was selected. A P-value <0.05 was considered statistically significant. US-guided biopsy revealed 65 benign lesions and 21 malignant lesions. The mean conductivity parameter of the single ROI method was selected (AUC = 0.74). Considering conductivity (≤0.10 S/m), ADC (≥1.60 × 10-3 mm2 /sec), and BI-RADS categories 2 or 3 reduced false-positive biopsies by 23% (15 of 65), 38% (25 of 65), and 43% (28 of 65), respectively, without missing malignant lesions. Conductivity imaging may show lower performance than DWI and abbreviated MRI in reducing unnecessary biopsies. 2 TECHNICAL EFFICACY: Stage 2.

Quantification of brain developmental trajectories based on histogram parameters at slice of the anterior and posterior horns of the lateral ventricles on routine brain MRI in normal children aged 0 to 5 years

Objective: To investigate the value of histogram parameters in quantifying brain development trajectory at slice of anterior and posterior horns of lateral ventricles on conventional brain MRI in normal children aged 0-5 years. Methods: Routine brain MRI data [apparent diffusion coefficient (ADC) map, T1-weighted imaging (T1WI), and T2-weighted imaging (T2WI)] were retrospectively collected from 300 children aged 0-5 years who underwent MRI at Children 's Hospital of Nanjing Medical University from April 2014 to November 2021, 154 males and 146 females, aged [M (Q1, Q3) ] 35.57(17.98,50.66)months. According to the random sampling method, they were divided into training set (n=240) and validation set (n=60) in a ratio of 8∶2. The training set was divided into 6 groups according to age:≤0.5 years, 24 persons; >0.5-≤1 years,21 persons; >1-≤2 years,31 persons; >2-≤3 years,44 persons; >3-≤4 years,42 persons; >4-≤5 years,78 persons. MRIcron software was used to delineate the whole brain at the level of the anterior and posterior horns of the lateral ventricles of the three MRI data as the region of interest. Then gray histograms and their parameters [including mean, maximum, minimum, skewness, kurtosis, mode, variance, and percentiles at 5% intervals from 10% to 95%(10th-95th) ]were obtained. Intra-class correlation coefficients (ICC) were used to assess consistency of intra-observer and inter-observer measurement. Representative parameters were selected by Spearman correlation analysis and curve fitting. The linear regression coefficient β represented development rates at different ages. The selected curve regression models were applied to the validation set, and the reliability of the model was evaluated with accuracy. Results: Intra-observer and inter-observer histogram measurement parameters were generally in good consistency (ICC>0.800, all P<0.001). Histogram parameters ADC 10th-65th, T1WI 55th-80th and T2WI 10th-45th were highly correlated with age (∣r∣≥0.700, 0.600 and 0.600 respectively; all P<0.001). ADC 30th and T2WI 10th had the greatest goodness of fit (R²=0.871, 0.873; both P<0.001). Map of brain development trends showed that ADC 30th and T2WI 10th decreased with age. ADC 30th changed rapidly before the age of 2 years, most significantly within 6 months, and the rate of decrease slowed down after 2 years old. T2WI 10th decreased rapidly within 1 year, and moderately after 1 year old. The curve regression models of ADC 30th and T2WI 10th had higher accuracy in validation set [93% (56/60) and 95% (57/60), respectively]. Conclusion: Histogram parameters can quantify brain developmental trajectories at slice of anterior and posterior horns of lateral ventricles on conventional MRI in normal children aged 0-5 years, and obtain the brain development curves reflecting this slice of this age group.

Differentiation of glioblastoma, brain metastases and central nervous system lymphomas using amount of vasogenic edema and diffusion MR imaging of tumor core and peritumoral zone- Searching for a practical approach

Background/Aim: The differential diagnosis of solitary brain tumors poses challenges for clinicians and radiologists, often leading to invasive biopsy procedures. Therefore, this study aimed to evaluate the variations in edema volume and diffusion characteristics between the tumor core and peritumoral zone in cases of glioblastoma, brain metastasis, and central nervous system lymphoma. The aim was to identify additional parameters for conventional magnetic resonance imaging (MRI) that could aid in the differential diagnosis. Methods: A total of 39 patients (13 with central nervous system lymphoma, 13 with glioblastoma, and 13 with brain metastases) were included in this retrospective cohort study. Apparent diffusion coefficient (ADC) values were calculated from the ADC maps obtained from Brain MRI for both the lesion and peritumoral region. Additionally, the largest diameter of the vasogenic edema-mass complex was measured using T2 sequences. In the contrast-enhanced series, the largest diameter of the metastatic lesion was measured. The edema-mass ratio was determined by dividing the diameter of the edema-mass complex by the diameter of the mass. Results: There was a statistically significant difference in the edema-mass ratio among the tumor types (P=0.008). Further analysis using Bonferroni correction revealed that this difference was primarily due to glioblastoma. Compared to patients with lymphoma and brain metastases, lesions diagnosed as glioblastoma exhibited a lower edema-mass ratio. Additionally, a statistically significant difference was observed in the ADC value measured from the lesion according to the tumor type (P=0.017). It was determined that lesions associated with central nervous system lymphoma had lower ADC values than those with glioblastoma. Conclusion: Including lesional and perilesional ADC values obtained through diffusion-weighted examination and edema mass ratio measurements may enhance the accuracy of differential diagnosis. Utilizing these imaging characteristics in a multiparametric approach, as suggested by this research, can improve the accuracy of diagnosing malignant cancers, thereby enabling better patient management and treatment decisions.

Classification of brain lesions using a machine learning approach with cross-sectional ADC value dynamics

Diffusion-weighted imaging (DWI) and its numerical expression via apparent diffusion coefficient (ADC) values are commonly utilized in non-invasive assessment of various brain pathologies. Although numerous studies have confirmed that ADC values could be pathognomic for various ring-enhancing lesions (RELs), their true potential is yet to be exploited in full. The article was designed to introduce an image analysis method allowing REL recognition independently of either absolute ADC values or specifically defined regions of interest within the evaluated image. For this purpose, the line of interest (LOI) was marked on each ADC map to cross all of the RELs’ compartments. Using a machine learning approach, we analyzed the LOI between two representatives of the RELs, namely, brain abscess and glioblastoma (GBM). The diagnostic ability of the selected parameters as predictors for the machine learning algorithms was assessed using two models, the k-NN model and the SVM model with a Gaussian kernel. With the k-NN machine learning method, 80% of the abscesses and 100% of the GBM were classified correctly at high accuracy. Similar results were obtained via the SVM method. The proposed assessment of the LOI offers a new approach for evaluating ADC maps obtained from different RELs and contributing to the standardization of the ADC map assessment.

3.0 T prostate MRI: Visual assessment of 2D and 3D T2-weighted imaging sequences using PI-QUAL score

PurposeComparing 2D and 3D T2 weighted sequences in terms of image quality in 3.0 T MRI with readers of varied experiences, using PI-QUAL inspired criteria. Methods91 male patients with suspected prostate cancer (PCa) underwent diagnostic prostate MRI on a 3.0 T MR system using a 32-channel phased-array torso coil before prostate biopsy. MRI protocol included 3D T2w images, axial 2D T2w images, axial diffusion-weighted images (DWI) with the corresponding ADC apparent diffusion coefficient maps, and axial dynamic contrast enhanced images. 3D T2w and 2D T2w imaging were compared by 4 radiologists using a Likert scale for image quality (overall anatomy, delineation of capsule, seminal vesicles, ejaculatory ducts, sphincter muscle, artifacts), tumor delimitation and conspicuity. ResultsNo significant differences in terms of overall quality between 3D and 2D T2w images were found. However 2D T2w demonstrated higher rating than 3D T2w images as for the image quality of the external capsule, sphincter muscle and ejaculatory ducts delineation (p < 0.05). Conclusion3D T2w sequence can’t replace 2D T2w sequence, despite good quality images but it remains more prone to artifacts. Quality of 2D T2w sequences was substantially superior to 3D sequences for delineation of key structures as external capsule, sphincter muscle. The use of PI-QUAL criteria allows reproducible analysis of the quality of T2 weighted images.

Diffusion-Weighted Images and Contrast-Enhanced MRI in the Diagnosis of Different Stages of Multiple Sclerosis of the Central Nervous System.

Introduction Multiple sclerosis (MS) is one of the most prevalent disorders of the central nervous system (CNS), and it can be observed in the field of radiological cross-sectional magnetic resonance imaging (MRI). The prevalence of MS in Saudi Arabia has increased as compared to the past few years. MRI is the gold standard non-invasive modality of choice in MS diagnosis according to the National Multiple Sclerosis Society (NMSS), New York City. This study aimed to highlight the significance of using diffusion-weighted images (DWIs) and the use of contrast media in the MS protocol, as well as the importance of identifying the suitable time of imaging after contrast enhancement to detect active lesions. Methods A retrospective cross-sectional study was conducted of 100 MS patients with an age range of 17 to 56 years. The data set included 41 active cases and 59 inactive cases. All patients had an MRI standard protocol of both the brain and spine in addition to DWI sequence and contrast agent (CA) injection, with images taken in early and delayed time. Results Of the patients, 71% were female and 29% were male. Active MS disease was more significant at younger ages than at older ages. Active lesions were significantly enhanced in delayed contrast images and showed high signal intensity in both the DWI and apparent diffusion coefficient (ADC) map, while inactive lesions showed no enhancement after contrast injection and showed an iso-signal intensity in both the DWI and ADC map. Conclusion The use of CA has developed over the years in the diagnosis of MS patients. In this study, the relationship between active lesions, DWI, and delayed contrast enhancement is very strong. In future research, we recommend adding a DWI sequence for the suspected active MS spine lesions in addition to delayed enhancement time in active MS after contrast injection to increase MRI sensitivity toward active MS lesions of the brain and spinal cord as well.

Role of Diffusion-Weighted MRI ( DWI) in Differentiating Between Benign and Malignant Nodules of Thyroid Taking Histopathology as Gold Standard

Background: Thyroid gland is a principal endocrine organ, first to develop in embryo, endodermal in origin and producing hormones essential for life. It is a vascular organ inclusive of right and left lobes, surrounded by a capsule and isthmus in the centre. Objective: To determine the role of Diffusion-Weighted MRI ( DWI) in differentiating between benign and malignant thyroid nodules keeping histopathology as the gold standard. Methodology: One hundred and three patients (103) were enrolled. DWI sequence was performed on 1.5 Tesla GE machine at b-values of 0, 50 and 1000 s/mm2 with correlative ADC map and quantitative values were calculated. FNAC of the thyroid nodules was carried out and results were tallied with ADC values. Out of these, five patients lost to follow up and an inadequate sample was obtained in six patients. Results: Mean age of patients (n=92) was 39 years. Out of 92 patients, 26 patients (28.3%) showed restricted diffusion on DWI in malignant thyroid nodules. Mean ADC ( Apparent Diffusion Coefficient) value of benign&#x0D; thyroid nodules (1.43 × 10-3mm2s-1) was significantly greater than malignant thyroid nodules (0.91×10-3mm s-1). For discriminating two types of nodules, cut off ADC value was determined at 1.1×10-3mm2s-1 and its sensitivity, specificity, PPV, NPV and diagnostic accuracy was 84.6, 95.4, 88.8, 91.5 and 92.3% respectively. Conclusion: Diffusion Weighted MRI is a non-invasive imaging investigation without ionizing radiation hazard. Its greater soft tissue differentiation and multiplanar images help in the diagnosis of malignant thyroid nodules with high sensitivity and accuracy. The overload of unnecessary surgeries is thus lowered when pre-operative FNAC is indecisive and aids in making a precise diagnosis.

Diffusion-weighted Imaging of the Chest: A Primer for Radiologists.

Diffusion-weighted imaging (DWI) is a fundamental sequence not only in neuroimaging but also in oncologic imaging and has emerging applications for MRI evaluation of the chest. DWI can be used in clinical practice to enhance lesion conspicuity, tissue characterization, and treatment response. While the spatial resolution of DWI is in the order of millimeters, changes in diffusion can be measured on the micrometer scale. As such, DWI sequences can provide important functional information to MRI evaluation of the chest but require careful optimization of acquisition parameters, notably selection of b values, application of parallel imaging, fat saturation, and motion correction techniques. Along with assessment of morphologic and other functional features, evaluation of DWI signal attenuation and apparent diffusion coefficient maps can aid in tissue characterization. DWI is a noninvasive noncontrast acquisition with an inherent quantitative nature and excellent reproducibility. The outstanding contrast-to-noise ratio provided by DWI can be used to improve detection of pulmonary, mediastinal, and pleural lesions, to identify the benign nature of complex cysts, to characterize the solid portions of cystic lesions, and to classify chest lesions as benign or malignant. DWI has several advantages over fluorine 18 (18F)-fluorodeoxyglucose PET/CT in the assessment, TNM staging, and treatment monitoring of lung cancer and other thoracic neoplasms with conventional or more recently developed therapies. © RSNA, 2023 Quiz questions for this article are available in the supplemental material. Supplemental material and the slide presentation from the RSNA Annual Meeting are available for this article.

An Apparent Diffusion Coefficient-Based Machine Learning Model Can Improve Prostate Cancer Detection in the Grey Area of the Prostate Imaging Reporting and Data System Category 3: A Single-Centre Experience

The Prostate Imaging and Reporting Data System (PI-RADS) has a key role in the management of prostate cancer (PCa). However, the clinical interpretation of PI-RADS 3 score lesions may be challenging and misleading, thus postponing PCa diagnosis to biopsy outcome. Multiparametric magnetic resonance imaging (mpMRI) radiomic analysis may represent a stand-alone noninvasive tool for PCa diagnosis. Hence, this study aims at developing a mpMRI-based radiomic PCa diagnostic model in a cohort of PI-RADS 3 lesions. We enrolled 133 patients with 155 PI-RADS 3 lesions, 84 of which had PCa confirmation by fusion biopsy. Local radiomic features were generated from apparent diffusion coefficient maps, and the four most informative were selected using LASSO, the Wilcoxon rank-sum test (p &lt; 0.001), and support vector machines (SVMs). The selected features where augmented and used to train an SVM classifier, externally validated on a holdout subset. Linear and second-order polynomial kernels were exploited, and their predictive performance compared through receiver operating characteristics (ROC)-related metrics. On the test set, the highest performance, equally for both kernels, was specificity = 76%, sensitivity = 78%, positive predictive value = 80%, and negative predictive value = 74%. Our findings substantially improve radiologist interpretation of PI-RADS 3 lesions and let us advance towards an image-driven PCa diagnosis.

Novel Multiparametric Magnetic Resonance Imaging-Based Deep Learning and Clinical Parameter Integration for the Prediction of Long-Term Biochemical Recurrence-Free Survival in Prostate Cancer after Radical Prostatectomy

Radical prostatectomy (RP) is the main treatment of prostate cancer (PCa). Biochemical recurrence (BCR) following RP remains the first sign of aggressive disease; hence, better assessment of potential long-term post-RP BCR-free survival is crucial. Our study aimed to evaluate a combined clinical-deep learning (DL) model using multiparametric magnetic resonance imaging (mpMRI) for predicting long-term post-RP BCR-free survival in PCa. A total of 437 patients with PCa who underwent mpMRI followed by RP between 2008 and 2009 were enrolled; radiomics features were extracted from T2-weighted imaging, apparent diffusion coefficient maps, and contrast-enhanced sequences by manually delineating the index tumors. Deep features from the same set of imaging were extracted using a deep neural network based on pretrained EfficentNet-B0. Here, we present a clinical model (six clinical variables), radiomics model, DL model (DLM-Deep feature), combined clinical–radiomics model (CRM-Multi), and combined clinical–DL model (CDLM-Deep feature) that were built using Cox models regularized with the least absolute shrinkage and selection operator. We compared their prognostic performances using stratified fivefold cross-validation. In a median follow-up of 61 months, 110/437 patients experienced BCR. CDLM-Deep feature achieved the best performance (hazard ratio [HR] = 7.72), followed by DLM-Deep feature (HR = 4.37) or RM-Multi (HR = 2.67). CRM-Multi performed moderately. Our results confirm the superior performance of our mpMRI-derived DL algorithm over conventional radiomics.

Peritumoral Radiomics Strategy Based on Ensemble Learning for the Prediction of Gleason Grade Group of Prostate Cancer

To develop and evaluate a peritumoralradiomic-based machine learning model to differentiate low-Gleason grade group (L-GGG) and high-GGG (H-GGG) prostate lesions. In this retrospective study, a total of 175 patients with prostate cancer (PCa) confirmed by puncture biopsy were recruited and included 59 patients with L-GGG and 116 patients with H-GGG. The original PCa regions of interest (ROIs) were delineated on T2-weighted (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps, and then centra-tumoral and peritumoral ROIs were defined. Features were meticulously extracted from each ROI to establish radiomics models, employing distinct sequence datasets. Peritumoral radiomics models were specifically developed for both the peripheral zone (PZ) and transitional zone (TZ), utilizing dedicated PZ and TZ datasets, respectively. The performances of the models were evaluated by using the receiver operating characteristic (ROC) curve and precision-recall curve. The classification model with combined peritumoral features based on T2+DWI+ADC sequence dataset demonstrated superior performance compared to the original tumor and centra-tumoral classification models. It achieved an area under the ROC curve (AUC) of 0.850 [95% confidence interval, 0.849, 0.860] and an average accuracy of 0.950. The combined peritumoral model outperformed the regional peritumoral models with AUC of 0.85 versus 0.75 for PZ lesions and 0.88 versus 0.69 for TZ lesions, respectively. The peritumoral classification models exhibit greater efficacy in predicting PZ lesions as opposed to TZ lesions. The peritumoral radiomics features showed excellent performance in predicting GGG in PCa patients and might be a valuable addition to the non-invasive assessment of PCa aggressiveness.

Evaluation of the efficacy of diffusion-weighted magnetic resonance imaging and apparent diffusion coefficients in the diagnosis of maxillary diseases

We investigated the efficacy of using diffusion-weighted imaging (DWI) with quantitative apparent diffusion coefficient (ADC) mapping in the diagnosis of maxillary diseases. We evaluated 146 cases of maxillary diseases (32 malignant tumors, 11 benign tumors, 28 maxillary cysts, 60 cases of maxillary sinusitis, and 15 maxillary sinus retention cysts) that had been examined using magnetic resonance imaging. The DWI sequence was obtained with b values of 0 and 800 s/mm2 and ADC values were calculated. We used one-way analysis of variance and the Tukey honestly significant difference test to identify differences within and between the types of diseases. Mean ADC values for malignant tumors (1.07 × 10-3 mm2 s-1) were significantly lower than ADCs for benign tumors (1.85 × 10-3 mm2 s-1), maxillary cysts (1.77 × 10-3 mm2 s-1), maxillary sinusitis (2.34 × 10-3 mm2 s-1), and maxillary sinus retention cysts (2.52 × 10-3 mm2 s-1), with P < 0.001. Mean ADC differed significantly between all disease types except between maxillary sinusitis and maxillary sinus retention cysts. ADC values also significantly differed between specific lesions within the malignant tumor and maxillary cyst groups. The use of ADC values can be useful in the differential diagnosis of malignant maxillary diseases, benign lesions, cysts, and inflammatory and reactive conditions.

Apparent diffusion coefficient correlates with different histopathological features in several intrahepatic tumors

ObjectivesTo investigate associations between apparent diffusion coefficient (ADC) and cell count, Ki 67, tumor-stroma ratio (TSR), and tumoral lymphocytes in different hepatic malignancies.MethodsWe identified 149 cases with performed liver biopsies: hepatocellular cancer (HCC, n = 53), intrahepatic cholangiocarcinoma (iCC, n = 29), metastases of colorectal cancer (CRC, n = 24), metastases of breast cancer (BC, n = 28), and metastases of pancreatic cancer (PC, n = 15). MRI was performed on a 1.5-T scanner with an axial echo-planar sequence. MRI was done before biopsy. Biopsy images of target lesions were selected. The cylindrical region of interest was placed on the ADC map of target lesions in accordance with the needle position on the biopsy images. Mean ADC values were estimated. TSR, cell counts, proliferation index Ki 67, and number of tumor-infiltrating lymphocytes were estimated. Spearman’s rank correlation coefficients and intraclass correlation coefficients were calculated.ResultsInter-reader agreement was excellent regarding the ADC measurements. In HCC, ADC correlated with cell count (r = − 0.68, p < 0.001) and with TSR (r = 0.31, p = 0.024). In iCC, ADC correlated with TSR (r = 0.60, p < 0.001) and with cell count (r = − 0.54, p = 0.002). In CRC metastases, ADC correlated with cell count (r = − 0.54 p = 0.006) and with Ki 67 (r = − 0.46, p = 0.024). In BC liver metastases, ADC correlated with TSR (r = 0.55, p < 0.002) and with Ki 67 (r = − 0.51, p = 0.006). In PC metastases, no significant correlations were found.ConclusionsADC correlated with tumor cellularity in HCC, iCC, and CRC liver metastases. ADC reflects TSR in BC liver metastases, HCC, and iCC. ADC cannot reflect intratumoral lymphocytes.Clinical relevance statementThe present study shows that the apparent diffusion coefficient can be used as a surrogate imaging marker for different histopathological features in several malignant hepatic lesions.Key Points• ADC reflects different histopathological features in several hepatic tumors.• ADC correlates with tumor cellularity in HCC, iCC, and CRC metastases.• ADC strongly correlates with tumor-stroma ratio in BC metastases and iCC.