Deep learning for assessing image quality in bi-parametric prostate MRI: A feasibility study

BACKGROUND

Although systems such as Prostate Imaging Quality (PI-QUAL) have been proposed for quality assessment, visual evaluations by human readers remain somewhat inconsistent, particularly among less-experienced readers.

OBJECTIVES

To assess the feasibility of deep learning (DL) for the automated assessment of image quality in bi-parametric MRI scans and compare its performance to that of less-experienced readers.

METHODS

We used bi-parametric prostate MRI scans from the PI-CAI dataset in this study. A 3-point Likert scale, consisting of poor, moderate, and excellent, was utilized for assessing image quality. Three expert readers established the ground-truth labels for the development (500) and testing sets (100). We trained a 3D DL model on the development set using probabilistic prostate masks and an ordinal loss function. Four less-experienced readers scored the testing set for performance comparison.

RESULTS

The kappa scores between the DL model and the expert consensus for T2W images and ADC maps were 0.42 and 0.61, representing moderate and good levels of agreement. The kappa scores between the less-experienced readers and the expert consensus for T2W images and ADC maps ranged from 0.39 to 0.56 (fair to moderate) and from 0.39 to 0.62 (fair to good).

CONCLUSIONS

Deep learning (DL) can offer performance comparable to that of less-experienced readers when assessing image quality in bi-parametric prostate MRI, making it a viable option for an automated quality assessment tool. We suggest that DL models trained on more representative datasets, annotated by a larger group of experts, could yield reliable image quality assessment and potentially substitute or assist visual evaluations by human readers.

Inter-reader agreement of the prostate imaging quality (PI-QUAL) score for basic readers in prostate MRI: A multi-center study

BACKGROUND

The Prostate Imaging Quality (PI-QUAL) score is the first step toward image quality assessment in multi-parametric prostate MRI (mpMRI). Previous studies have demonstrated moderate to excellent inter-rater agreement among expert readers; however, there is a need for studies to assess the inter-reader agreement of PI-QUAL scoring in basic prostate readers.

OBJECTIVES

To assess the inter-reader agreement of the PI-QUAL score amongst basic prostate readers on multi-center prostate mpMRI.

METHODS

Five basic prostate readers from different centers assessed the PI-QUAL scores independently using T2-weighted images, diffusion-weighted imaging (DWI) including apparent diffusion coefficient (ADC) maps, and dynamic-contrast-enhanced (DCE) images on mpMRI data obtained from five different centers following Prostate Imaging-Reporting and Data System Version 2.1. The inter-reader agreements amongst radiologists for PI-QUAL were evaluated using weighted Cohen's kappa. Further, the absolute agreements in assessing the diagnostic adequacy of each mpMRI sequence were calculated.

RESULTS

A total of 355 men with a median age of 71 years (IQR, 60-78) were enrolled in the study. The pair-wise kappa scores ranged from 0.656 to 0.786 for the PI-QUAL scores, indicating good inter-reader agreements between the readers. The pair-wise absolute agreements ranged from 0.75 to 0.88 for T2W imaging, from 0.74 to 0.83 for the ADC maps, and from 0.77 to 0.86 for DCE images.

CONCLUSIONS

Basic prostate radiologists from different institutions provided good inter-reader agreements on multi-center data for the PI-QUAL scores.

Does deep learning software improve the consistency and performance of radiologists with various levels of experience in assessing bi-parametric prostate MRI?

OBJECTIVE

To investigate whether commercially available deep learning (DL) software improves the Prostate Imaging-Reporting and Data System (PI-RADS) scoring consistency on bi-parametric MRI among radiologists with various levels of experience; to assess whether the DL software improves the performance of the radiologists in identifying clinically significant prostate cancer (csPCa).

METHODS

We retrospectively enrolled consecutive men who underwent bi-parametric prostate MRI at a 3 T scanner due to suspicion of PCa. Four radiologists with 2, 3, 5, and > 20 years of experience evaluated the bi-parametric prostate MRI scans with and without the DL software. Whole-mount pathology or MRI/ultrasound fusion-guided biopsy was the reference. The area under the receiver operating curve (AUROC) was calculated for each radiologist with and without the DL software and compared using De Long's test. In addition, the inter-rater agreement was investigated using kappa statistics.

RESULTS

In all, 153 men with a mean age of 63.59 ± 7.56 years (range 53-80) were enrolled in the study. In the study sample, 45 men (29.80%) had clinically significant PCa. During the reading with the DL software, the radiologists changed their initial scores in 1/153 (0.65%), 2/153 (1.3%), 0/153 (0%), and 3/153 (1.9%) of the patients, yielding no significant increase in the AUROC (p > 0.05). Fleiss' kappa scores among the radiologists were 0.39 and 0.40 with and without the DL software (p = 0.56).

CONCLUSIONS

The commercially available DL software does not increase the consistency of the bi-parametric PI-RADS scoring or csPCa detection performance of radiologists with varying levels of experience.

Is there any additional benefit of 68Ga-PSMA PET on radiotherapy target volume definition in patients with glioblastoma?

OBJECTIVE

To investigate the contribution of 68Gallium (68Ga)-PSMA (prostate-specific membrane antigen) positron emission tomography (PET) in defining radiotherapy (RT) target volume for glioblastoma and to compare the target volumes defined by Magnetic Resonance Imaging (MRI).

METHODS

RT planning Computed Tomography (CT) images were fused separately with pre-operative MRI and PET/MRI images of 10 glioblastoma patients, retrospectively. The contrast-enhanced area in T1 weighted MRI was contoured as gross tumor volume (GTV) and clinical target volume (CTV1) was obtained by including the cavity and T2/FLAIR hyperintense areas after giving a margin of 2 cm to the GTV. 68Ga-PSMA uptake area was contoured as biological tumor volume (BTV) and CTV2 was obtained with a margin of 2 cm to BTV. Planning target volumes (PTVs) were created with the 3 mm added to the CTVs. Conformity index (CI), dice similarity coefficient (DSC) and overlap volume (OV) were calculated by obtaining the intersection and union volumes. Volumetric comparison, similarity and overlap analyzes were performed statistically by Wilcoxon signed rank and One sample t-test.

RESULTS

The median GTV was 21,96 cc (1,04 - 82,04) and BTV was 25,58 cc (2,43 - 99,47). BTV was on average 47% larger than GTV which was statistically significant (p = 0.03). For GTV-BTV, CTV1-CTV2 and PTV1-PTV2; mean values of CI were 0,56, 0,76 and 0,76; DSC were 0,70, 0,86 and 0,86; OV were 0,88, 0,94 and 0,94, respectively. There was no significant difference on size and spatial similarity between CTV1 and CTV2, PTV1 and PTV2.

CONCLUSION

Altough BTV was larger than GTV, this significance was lost while we gave the same CTV margin including the peripheral edema. It seems that it may help to improve defining non-enhancing tumor part and also recurrent tumor volume.

ADVANCES IN KNOWLEDGE

Recent studies have focused on the role of 68Ga-PSMA PET in imaging of glial tumors. It has been observed that 68Ga-PSMA PET can clearly define the tumor borders and it can be beneficial in target volume delineation, especially in reirradiation of recurrent tumors.

The Diagnostic Value of Superb Microvascular Imaging in Prediction of Uterine Artery Embolization Treatment Response in Uterine Leiomyomas

OBJECTIVES

We aimed to determine if superb microvascular imaging (SMI) can predict response to uterine artery embolization (UAE) as compared with power Doppler ultrasound.

METHODS

The blood flow and the volume of the dominant leiomyoma was evaluated by power Doppler ultrasonography (PDUS) and SMI 1 day before and 3 months after the UAE procedure. SMI and PDUS blood flow were classified to 4 grades of vascularity. The change in fibroid volume in Grades 0-2 (hypovascular group) was compared to the hypervascular Grade 3 group.

RESULTS

Twenty-eight women (mean age, 40.9 years; range, 33-53 years) were examined with PDUS and SMI before and 3 months after UAE. The volume reduction was statistically significantly higher hypervascular group (P < .05). When we accept 30% or more volume reduction as a good response to UAE, the positive predictive value, negative predictive value, sensitivity, specificity, and accuracy of SMI were 100, 64, 73.6, 100, and 82.1%, respectively. There was excellent agreement between the two blinded observers in SMI measurements.

CONCLUSIONS

SMI, with its high reproducibility, provides further microvessel information than PDUS in uterine fibroids. It may be a useful tool in prediction of response to UAE treatment and improve counseling and patient selection for UAE versus medical or surgical treatment options.

Focusing on Asthma and Chronic Obstructive Pulmonary Disease with COVID-19

INTRODUCTION

We aimed to evaluate clinical and laboratory findings of hospitalized asthma and chronic obstructive pulmonary disease (COPD) patients with COVID-19 and demonstrate that they have different symptoms and/or laboratory results and outcomes than COVID-19 patients with comorbidity (CoV-com) and without comorbidity (CoV-alone).

METHODOLOGY

The data of the demographic, clinical, laboratory findings of hospitalized CoV-alone, asthma, COPD patients with COVID-19 (CoV-asthma, CoV-COPD, respectively), and CoV-com were analyzed.

RESULTS

Out of 1082 patients hospitalized for COVID-19, 585 (54.1%) had CoV-alone, 40 (3.7%) had CoV-asthma, 46 (4.3%) had CoV-COPD and 411 (38%) had CoV-com. Cough, shortness of breath, fever and weakness were the most common four symptoms seen in all COVID-19 patients. Shortness of breath, myalgia, headache symptoms were more common in CoV-asthma than the other groups (p < 0.001, p < 0.01, p < 0.05 respectively). Sputum was more common in CoV-COPD than other groups (p < 0.01). COPD group most frequently had increased values, different from the other groups with CRP>5ng/mL in 91.3%, D-dimer > 0.05mg/dL in 89.1%, troponin > 0.014micg/L in %63.9, INR>1.15 in 52.2%, CK-MB>25U/L in 48.5%, PT>14s in 40.9% of patients (p < 0.05, p < 0.001, p < 0.001, p < 0.001, p < 0.05, p < 0.001, respectively). NT-ProBNP was found to have the highest AUC value and the best differentiating parameter for CoV-asthma from CoV-alone. Typical CT findings were present in 44.4% of CoV-alone, 57.5% of CoV-asthma, 28.3% of CoV-COPD and 38.9% of CoV-com groups. CoV-COPD and CoV-com patients died more frequently than other groups (17.8%, 18.5%).

CONCLUSIONS

CoV-asthma and CoV-COPD patients might have different symptoms and laboratory parameters than other COVID-19 patients which can guide the physicians.

Fibroblast Growth Factor Receptor-4 Expression in Pituitary Adenomas is Associated with Aggressive Tumor Features

PURPOSE

To investigate the relationship of Fibroblast Growth Factor Receptor-4 (FGFR-4) expression with radiologic, pathologic, and clinical parameters in pituitary adenomas.

METHODS

Among 307 patients who underwent pituitary surgery for a pituitary adenoma between 2000 and 2015, we included 161 patients (53 gonadotroph, 26 corticotroph, 25 null cell, 22 lactotroph, 13 somatotroph, 8 adenomas with unusual combination, 7 Pit-1 positive adenomas, and 7 lactosomatotroph) based on availability of pathology specimens. Patients' radiologic, pathologic, and clinical parameters were determined. FGFR-4 immunostaining was evaluated using a semi-quantitative histologic score (H-score).

RESULTS

The mean follow-up period was 61 (IQR=32-84) months. The median H-scores for FGFR-4 were higher in patients without remission, those with residual lesion, and T2-hyperintense adenoma (p<0.05). Ki-67 level was higher in patients without remission compared to those in remission (p<0.05). The mean Ki-67 levels did not differ between patients with and without residual lesion or T2-hyperintense tumor (p>0.05). There was no significant difference (p>0.05) when the H-score and Ki-67 levels were assessed in terms of sex, sellar-dural invasion, Knosp and a grading system for superior, inferior, parasellar, anterior and posterior tumor extension Classification, tumor function or presence of poor subtype. Adenomas with Ki-67 expression ≥3% had higher FGFR4 expression levels than those with <3% expression (p=0.002). There was a weak positive correlation between H-score and Ki-67 (p=0.011; r=0.201).

CONCLUSIONS

Higher levels of FGFR-4 in pituitary adenomas could be use a marker for more aggressive tumor behavior.

MON-269 FGFR-4 Expression in Pituitary Adenomas Is Associated with Aggressive Tumor Features

Biomarkers predicting tumor aggressiveness in pituitary adenomas have been largely investigated, albeit, with inconsistent results. We investigated the relationship of Fibroblast Growth Factor Receptor-4 (FGFR-4) expression and determined its relationship with radiological, pathological, and clinical parameters.

In our study, 650 patients who were followed up for pituitary disease were reviewed from medical charts retrospectively. Of the 307 patients who underwent pituitary surgery for a pituitary adenoma between 2000 and 2015, we selected 161 cases on the basis of availability of pathology specimen of hypophysis tissue in our center. Patients’ radiological, pathological, and clinical parameters were obtained from medical charts. FGFR-4 immunostaining was evaluated using a semiquantitative Histologic score (H score).

The mean age of the patients was 56.02 ± 14.80 years. Ninety-two (57.1%) were male and 69 (42.9%) were female. The mean follow-up period was 68.94 ± 44.15 months. Pathological examination revealed the following subtypes; 53 nonfunctioning pituitary adenomas, 26 corticotroph adenomas, 25 hormone receptor negative adenomas, 22 mammotroph adenomas, 13 somatotroph adenomas, 8 combined hormone secreting adenomas, 7 somatomammotroph, and 7 PIT-1 positive adenomas. The mean tumor size was 26.83 ± 14.92 mm. In patients with cavernous sinus invasion, the mean adenoma size was significantly higher than those without (p <0.001). Mean H-score and Ki-67 levels were not different between patients with and without cavernous sinus invasion (p>0.05 for all). The mean H-score, Ki-67, and adenoma size were significantly higher in patients without remission than those with remission (p <0.001, p = 0.014, p <0.001, respectively). The mean H score and adenoma size were significantly higher in patients with residual lesions than those without (p = 0.002, p <0.001; respectively); there was no significant difference in Ki-

67 levels (p>0.05). When the H-score and Ki-67 levels were assessed in terms of gender, sellar-dural invasion, tumor function or presence of poor subtype, no significant difference was detected (p>0.05 for all). The mean H score and adenoma size were significantly higher in patients with Ki-67≥ 3 than those with <3 (p = 0.002, p = 0.004; respectively). There was a weak positive correlation between H-score and Ki-67 (p = 0.005; r = 0.218); on the other hand, Ki-67 was not correlated with mitosis grade, p53 staining, and age, respectively (p>0.05 for all).

In our study, we demonstrated that patients with residual lesion and those without remission had high expression of FGFR-4. Also FGFR-4 levels were positively correlated with Ki-67 which itself correlated with lack of remission. Taken together, our results indicate that higher levels of FGFR-4 and Ki-67 in pituitary adenomas might indicate a more aggressive tumor phenotype.

Penile masses: Shear Wave Elastography correlated with Magnetic Resonance Imagining. A two cases report

Definitive diagnosis of penile masses usually depends on clinical symptoms and patient history but in some challenging cases the help of radiologic diagnostic tools is required. Although ultrasound is an effective tool for detecting penile masses, unlike magnetic resonance imagining, it is not able to differentiate benign and malignant pathologies. Shear wave elastography (SWE) is a novel method, which has the ability to differentiate malignant and benign tissues by giving quantitative information about tissue elasticity. In this paper we present two cases with penile masses to demonstrate the potential use of SWE in the differential diagnosis of penile masses.