Deep-Learning Models for Detection and Localization of Visible Clinically Significant Prostate Cancer on Multi-Parametric MRI

BACKGROUND

Deep learning for diagnosing clinically significant prostate cancer (csPCa) is feasible but needs further evaluation in patients with prostate-specific antigen (PSA) levels of 4-10 ng/mL.

PURPOSE

To explore diffusion-weighted imaging (DWI), alone and in combination with T2-weighted imaging (T2WI), for deep-learning-based models to detect and localize visible csPCa.

STUDY TYPE

Retrospective.

POPULATION

One thousand six hundred twenty-eight patients with systematic and cognitive-targeted biopsy-confirmation (1007 csPCa, 621 non-csPCa) were divided into model development (N = 1428) and hold-out test (N = 200) datasets.

FIELD STRENGTH/SEQUENCE

DWI with diffusion-weighted single-shot gradient echo planar imaging sequence and T2WI with T2-weighted fast spin echo sequence at 3.0-T and 1.5-T.

ASSESSMENT

The ground truth of csPCa was annotated by two radiologists in consensus. A diffusion model, DWI and apparent diffusion coefficient (ADC) as input, and a biparametric model (DWI, ADC, and T2WI as input) were trained based on U-Net. Three radiologists provided the PI-RADS (version 2.1) assessment. The performances were determined at the lesion, location, and the patient level.

STATISTICAL TESTS

The performance was evaluated using the areas under the ROC curves (AUCs), sensitivity, specificity, and accuracy. A P value <0.05 was considered statistically significant.

RESULTS

The lesion-level sensitivities of the diffusion model, the biparametric model, and the PI-RADS assessment were 89.0%, 85.3%, and 90.8% (P = 0.289-0.754). At the patient level, the diffusion model had significantly higher sensitivity than the biparametric model (96.0% vs. 90.0%), while there was no significant difference in specificity (77.0%. vs. 85.0%, P = 0.096). For location analysis, there were no significant differences in AUCs between the models (sextant-level, 0.895 vs. 0.893, P = 0.777; zone-level, 0.931 vs. 0.917, P = 0.282), and both models had significantly higher AUCs than the PI-RADS assessment (sextant-level, 0.734; zone-level, 0.863).

DATA CONCLUSION

The diffusion model achieved the best performance in detecting and localizing csPCa in patients with PSA levels of 4-10 ng/mL.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Extracorporeal shock wave lithotripsy for treatment of large pediatric renal pelvic stone burden more than 2 cm

BACKGROUND

The high recurrence rates in pediatric urolithiasis indicate the need for none invasive or a minimally invasive treatment such as SWL. Therefore, EAU, ESPU and AUA recommend SWL as a first line treatment for renal calculi ≤ 2, and RIRS or PCNL for renal calculi > 2 cm. SWL is superior to RIRS and PCNL as it is inexpensive, outpatient procedure, and it has a high SFR in well selected cases specially pediatrics. On the other hand, SWL therapy has a limited efficacy with a lower SFR, and high retreatment rate and/or additional interventions for treatment of larger and harder renal calculi.

OBJECTIVE

We carried out this study to evaluate the efficacy and safety of SWL for treatment of renal stones > 2 cm to extend its indications for pediatric renal calculi.

METHODS

Between January 2016 and April 2022, we reviewed the records of patients with renal calculi treated by SWL, mini-PCNL, RIRS and open surgery in our institution. Forty-nine eligible children aged 1-5 years old, presented with renal pelvic and/or calyceal calculi measuring 2-3.9 cm and underwent SWL therapy were picked up and participated in the study. The data of an additional eligible 79 children with the same age and had renal pelvic and/or calyceal calculi > 2 cm up to stag horn calculi and underwent mini-PCNL, RIRS and open renal surgery were also picked up and participated in the study. We retrieved the following preoperative data from the records of the eligible patients; age, gender, weight, length, radiological findings (stone size, side, site, number and radio-density), renal function tests, routine laboratory findings, and urine analysis. The outcomes data in the form of; operative time, fluoroscopy time, hospital stay, SFRs, retreatment rates and complication rates were also retrieved from the records of patients treated with SWL and other techniques. Also, we collected the SWL characteristics in terms of; position, number and frequency of shocks, voltage, time of the session and U/S monitoring to assess stone fragmentation. All SWL procedures were performed according to the institution's standards.

RESULTS

The mean age of patients treated with SWL was 3.23 ± 1.19 years old, the mean size of the treated calculi was 2.31 ± 0.49 and the mean length of the SSD was 8.2 ± 1.4 cm. All patients had NCCT scan and the mean radio-density of the treated calculi was 572 ± 169.08 HUs based on NCCT scans Table (1). Single- and two-session SFRs of SWL therapy were 75.5% (37/49 patients) and 93.9% (46/49 patients), respectively. The overall success rate was 95.9% (47/49 patients) after three-session of SWL. Complications experienced by 7 patients (14.3%) in the form of fever (4.1%), vomiting (4.1%), abdominal pain (4/1%), and hematuria (2%). All complications were managed in outpatient settings. Our results were obtained on the basis of preoperative NCCT scans for all patients and postoperative plain KUB films and real-time abdominal U/S. Furthermore, single-session SFRs for SWL, mini-PCNL, RIRS and open surgery were 75.5%, 82.1%, 73.7% and 90.6%, respectively. Two-session SFRs by the same technique were 93.9%, 92.8%, and 89.5% for SWL, mini-PCNL and RIRS, respectively. A lower overall complication rate and higher overall SFR were found with SWL therapy compared to other techniques, Fig. (1).

DISCUSSION

Being a non-invasive outpatient procedure with a low complication rate and good spontaneous passage of stone fragments is the main advantage of SWL. In this study, the overall SFR is 93.9% where 46 out of 49 patients were completely rendered stone free after three session of SWL with overall success rate 95.9%. Badawy et al. reported overall success rates of 83.4% for renal stones with a mean stone size of 12.5 ± 7.2 mm. In children with renal stones measuring 18.2 mm, Ramakrishnan et al. reported a 97% SFR in accordance with our results. The high overall success rate (95.9%) and SFR (93.9%) in our research were attributed to the regular use of ramping procedure, low shock wave rate, percussion diuretics inversion (PDI) approach and alpha blocker therapy in all participants and short SSD. The limitations of our study are small sample of patients and its retrospective nature.

CONCLUSION

The non-invasive nature and replicability of the SWL procedure, along with the high success and low complication rates, give us a new insight to consider its application for treating pediatric renal calculi > 2 cm over the other more invasive techniques. Short SSD, the use of ramping procedure, low shock wave rate, 2 min break, PDI approach and alpha blockers therapy help better success of SWL.

LEVEL OF EVIDENCE

IV.

Time Course Changes of Synthetic Relaxation Time During Neoadjuvant Chemotherapy in Breast Cancer: The Optimal Parameter for Treatment Response Evaluation

BACKGROUND

Synthetic MRI (syMRI) has enabled quantification of multiple relaxation parameters (T1/T2 relaxation time [T1/T2], proton density [PD]), and their longitudinal change during neoadjuvant chemotherapy (NAC) promises to be valuable parameters for treatment response evaluation in breast cancer.

PURPOSE

To investigate the time course changes of syMRI parameters during NAC and evaluate their value as predictors for pathological complete response (pCR) in breast cancer.

STUDY TYPE

Retrospective, longitudinal.

POPULATION

A total of 129 women (median age, 50 years; range, 28-69 years) with locally advanced breast cancer who underwent NAC; all performed multiple conventional breast MRI examinations with added syMRI during NAC.

FIELD STRENGTH/SEQUENCE

A 3.0 T, T1-weighted dynamic contrast enhanced and syMRI acquired by a multiple-dynamic, multiple-echo sequence.

ASSESSMENT

Breast MRI was set at four time-points: baseline, after one cycle, after three or four cycles of NAC and preoperation. SyMRI parameters and tumor diameters were measured and their changes from baseline were calculated. All parameters were compared between pCR and non-pCR. Interaction between syMRI parameters and clinicopathological features was analyzed.

STATISTICAL TESTS

Mann-Whitney U tests, random effects model of repeated measurement, receiver operating characteristic (ROC) analysis, interaction analysis.

RESULTS

Median synthetic T1/T2/PD and tumor diameter generally decreased throughout NAC. Absolute T1 at early-NAC, T1, and PD at mid-NAC were significantly lower in the pCR group. After early-NAC, the T1 change was significantly higher in the pCR (median ± IQR, 18.17 ± 11.33) than the non-pCR group (median ± IQR, 10.90 ± 10.03), with the highest area under the ROC curves (AUC) of 0.769 (95% CI, 0.684-0.838). Interaction analysis showed that histological grade III patients had higher odds ratio (OR) (OR = 1.206) compared to grade II patients (OR = 1.067).

DATA CONCLUSION

Synthetic T1 changes after one cycle of NAC maybe useful for early evaluating NAC response in breast cancer during whole treatment cycles. However, its discriminative ability is significantly affected by histological grade.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 2.

Attention-guided multi-scale learning network for automatic prostate and tumor segmentation on MRI

BACKGROUND AND OBJECTIVE

Image-guided clinical diagnosis can be achieved by automatically and accurately segmenting prostate and prostatic cancer in male pelvic magnetic resonance imaging (MRI) images. For accurate tumor removal, the location, number, and size of prostate cancer are crucial, especially in surgical patients. The morphological differences between the prostate and tumor regions are small, the size of the tumor is uncertain, the boundary between the tumor and surrounding tissue is blurred, and the classification that separates the normal region from the tumor is uneven. Therefore, segmenting prostate and tumor on MRI images is challenging.

METHODS

This study offers a new prostate and prostatic cancer segmentation network based on double branch attention driven multi-scale learning for MRI. To begin, the dual branch structure provides two input images with different scales for feature coding, as well as a multi-scale attention module that collects details from different scales. The features of the double branch structure are then entered into the built feature fusion module to get more complete context information. Finally, to give a more precise learning representation, each stage is built using a deep supervision mechanism.

RESULTS

The results of our proposed network's prostate and tumor segmentation on a variety of male pelvic MRI data sets show that it outperforms existing techniques. For prostate and prostatic cancer MRI segmentation, the dice similarity coefficient (DSC) values were 91.65% and 84.39%, respectively.

CONCLUSIONS

Our method maintains high correlation and consistency between automatic segmentation results and expert manual segmentation results. Accurate automatic segmentation of prostate and prostate cancer has important clinical significance.

Endoscopic stone management in an ectopic ureter inserting into the prostatic urethra

We report a case of successful endoscopic stone management in a patient with ectopic ureteric insertion. The patient had a complete duplex collecting system, with the upper moiety ureter inserting ectopically into his prostatic urethra, and an obstructing ureteric stone in the distal portion of the ectopic ureter. This made both characterisation of the patient's anatomy and initial emergency stone management challenging.The case offers several learning points for clinicians who may encounter similar situations. By describing the challenges of managing this patient's presentation, we highlight considerations in imaging interpretation and operative approach that may help the reader manage a similar presentation to their practice. Additionally, we remind the urologist to consider the implications of an ectopic duplex ureter on future procedures, such as transurethral resection of the prostate or radical prostatectomy.

Quantitative dynamic contrast-enhance MRI parameters for rectal carcinoma characterization: correlation with tumor tissue composition

OBJECTIVE

To investigate the relationship between dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) measurements and the potential composition of rectal carcinoma.

METHODS

Twenty-four patients provided informed consent for this study. DCE-MRI was performed before total mesorectal excision. Quantitative parameters were calculated based on a modified Tofts model. Whole-mount immunohistochemistry and Masson staining sections were generated and digitized at histological resolution. The percentage of tissue components area was measured. Pearson correlation analysis was used to evaluate the correlations between pathological parameters and DCE-MRI parameters.

RESULTS

On the World Health Organization (WHO) grading scale, there were significant differences in extracellular extravascular space (Ktrans) (F = 9.890, P = 0.001), mean transit time (MTT) (F = 9.890, P = 0.038), CDX-2 (F = 4.935, P = 0.018), and Ki-67 (F = 4.131, P = 0.031) among G1, G2, and G3. ECV showed significant differences in extramural venous invasion (t =  - 2.113, P = 0.046). Ktrans was strongly positively correlated with CD34 (r = 0.708, P = 0.000) and moderately positively correlated with vimentin (r = 0.450, P = 0.027). Interstitial volume (Ve) was moderately positively correlated with Masson's (r = 0.548, P = 0.006) and vimentin (r = 0.417, P = 0.043). There was a moderate negative correlation between Ve and CDX-2 (r =  - 0.441, P = 0.031). The rate constant from extracellular extravascular space to blood plasma (Kep) showed a strong positive correlation with CD34 expression (r = 0.622, P = 0.001). ECV showed a moderate negative correlation with CDX-2 (r =  - 0.472, P = 0.020) and a moderate positive correlation with collagen fibers (r = 0.558, P = 0.005).

CONCLUSION

The dynamic contrast-enhanced MRI-derived parameters measured in rectal cancer were significantly correlated with the proportion of histological components. This may serve as an optimal imaging biomarker to identify tumor tissue components.

BBB opening by low pulsed electric fields, depicted by delayed-contrast MRI, enables efficient delivery of therapeutic doxorubicin doses into mice brains

BACKGROUND

Pharmacological treatment of CNS diseases is limited due to the presence of the blood-brain barrier (BBB). Recent years showed significant advancement in the field of CNS drug delivery enablers, with technologies such as MR-guided focused ultrasound reaching clinical trials. This have inspired researchers in the field to invent novel brain barriers opening (BBo) technologies that are required to be simple, fast, safe and efficient. One such technology, recently developed by us, is BDF (Barrier Disrupting Fields), based on low pulsed electric fields (L-PEFs) for opening the BBB in a controlled, safe, reversible and non-invasive manner. Here, we conducted an in vivo study to show that BDF is a feasible technology for delivering Doxorubicin (Doxo) into mice brain. Means for depicting BBBo levels were developed and applied for monitoring the treatment and predicting response. Overall, the goals of the presented study were to demonstrate the feasibility for delivering therapeutic Doxo doses into naïve and tumor-bearing mice brains and applying delayed-contrast MRI (DCM) for monitoring the levels of BBBo.

METHODS

L-PEFs were applied using plate electrodes placed on the intact skull of naïve mice. L-PEFs/Sham mice were scanned immediately after the procedure by DCM ("MRI experiment"), or injected with Doxo and Trypan blue followed by delayed (4 h) perfusion and brain extraction ("Doxo experiment"). Doxo concentrations were measured in brain samples using confocal microscopy and compared to IC50 of Doxo in glioma cell lines in vitro. In order to map BBBo extent throughout the brain, pixel by pixel MR image analysis was performed using the DCM data. Finally, the efficacy of L-PEFs in combination with Doxo was tested in nude mice bearing intracranial human glioma tumors.

RESULTS

Significant amount of Doxo was found in cortical regions of all L-PEFs-treated mice brains (0.50 ± 0.06 µg Doxo/gr brain) while in Sham brains, Doxo concentrations were below or on the verge of detection limit (0.03 ± 0.02 µg Doxo/gr brain). This concentration was x97 higher than IC50 of Doxo calculated in gl261 mouse glioma cells and x8 higher than IC50 of Doxo calculated in U87 human glioma cells. DCM analysis revealed significant BBBo levels in the cortical regions of L-PEFs-treated mice; the average volume of BBBo in the L-PEFs-treated mice was x29 higher than in the Sham group. The calculated BBBo levels dropped exponentially as a function of BBBo threshold, similarly to the electric fields distribution in the brain. Finally, combining non-invasive L-PEFs with Doxo significantly decreased brain tumors growth rates in nude mice.

CONCLUSIONS

Our results demonstrate significant BBBo levels induced by extra-cranial L-PEFs, enabling efficient delivery of therapeutic Doxo doses into the brain and reducing tumor growth. As BBBo was undetectable by standard contrast-enhanced MRI, DCM was applied to generate maps depicting the BBBo levels throughout the brain. These findings suggest that BDF is a promising technology for efficient drug delivery into the brain with important implications for future treatment of brain cancer and additional CNS diseases.

Experience of high polymer gel pad assisted ultrasound monitoring in the treatment of infant urolithiasis during extracorporeal shock wave lithotripsy

In the extracorporeal shock wave lithotripsy for infants, we used a medical polymer gel pad to assist ultrasonic positioning, so that the ultrasonic probe could be far away from the shock wave energy field. Although not affecting the ultrasonic positioning and monitoring effect, we discussed the protective effect of this method on the ultrasonic probe. A retrospective analysis was made on 21 infants (0-3 years old) who received ESWL in our hospital from June 2021 to February 2023. After the stones were accurately located by B-ultrasound before surgery, a 4 * 5 * 10 cm medical polymer gel pad was placed between the skin and the ultrasonic probe to keep the ultrasonic probe away from the shock wave energy field. The B-ultrasonic wave source locked the target stone through the gel pad, and the lithotripter Dornier Compact Delta II was used for lithotripsy. The extracorporeal shock wave lithotripsy was completed under the whole process of B-ultrasonic monitoring. All patients completed the surgery under ultrasound monitoring, and there were no abnormalities in the ultrasound probe during the surgery. The average stone size was 0.60 ± 0.21 cm, the surgical time was 39.8 ± 13.8 min, and the total energy of lithotripsy was 7.41 ± 4.35 J. There were no obvious complications in all patients after the surgery. After 2 weeks of ultrasound examination, the success rate of lithotripsy in 21 patients reached 85.7%. We believe that the use of the gel pad increases the distance between the ultrasonic probe and the skin, leaving the probe away from the shock wave energy field, avoiding the damage of the shock wave source to the ultrasonic probe, and does not affect the monitoring effect of ultrasound on stones and the success rate of lithotripsy, which is worthy of further promotion in the field of children's urinary stones.

A retrospective survey to establish institutional diagnostic reference levels for CT urography examinations based on clinical indications: preliminary results

Objective. To establish institutional diagnostic reference levels (IDRLs) based on clinical indications (CIs) for three- and four-phase computed tomography urography (CTU).Methods. Volumetric computed tomography dose index (CTDIvol), dose-length product (DLP), patients' demographics, selected CIs like lithiasis, cancer, and other diseases, and protocols' parameters were retrospectively recorded for 198 CTUs conducted on a Toshiba Aquilion Prime 80 scanner. Patients were categorised based on CIs and number of phases. These groups' 75th percentiles of CTDIvoland DLP were proposed as IDRLs. The mean, median and IDRLs were compared with previously published values.Results. For the three-phase protocol, the CTDIvol(mGy) and DLP (mGy.cm) were 22.7/992 for the whole group, 23.4/992 for lithiasis, 22.8/1037 for cancer, and 21.2/981 for other diseases. The corresponding CTDIvol(mGy) and DLP (mGy.cm) values for the four-phase protocol were 28.6/1172, 30.6/1203, 27.3/1077, and 28.7/1252, respectively. A significant difference was found in CTDIvoland DLP between the two protocols, among the phases of three-phase (except cancer) and four-phase protocols (except DLP for other diseases), and in DLP between the second and third phases (except for cancer group). The results are comparable or lower than most studies published in the last decade.Conclusions. The CT technologist must be aware of the critical dose dependence on the scan length and the applied exposure parameters for each phase, according to the patient's clinical background and the corresponding imaging anatomy, which must have been properly targeted by the competent radiologist. When clinically feasible, restricting the number of phases to three instead of four could remarkably reduce the patient's radiation dose. CI-based IDRLs will serve as a baseline for comparison with CTU practice in other hospitals and could contribute to national DRL establishment. The awareness and knowledge of dose levels during CTU will prompt optimisation strategies in CT facilities.

Contrast-enhanced Ultrasound Combined With Elastography for the Evaluation of Muscle-invasive Bladder Cancer in Rats

OBJECTIVES

By comparing with the control group, we evaluated the usefulness of contrast-enhanced ultrasound (CEUS) combined with elastography for the assessment of muscle invasion by bladder cancer (MIBC) in a Sprague-Dawley (SD) rat model.

METHODS

In the experimental group, 40 SD rats developed in situ bladder cancer (BLCA) in response to N-methyl-N-nitrosourea treatment, whereas 40 SD rats were included in the control group for comparison. We compared PI, Emean , microvessel density (MVD), and collagen fiber content (CFC) between the two groups. In the experimental group, Bland-Altman test was used to assess the relationships between various parameters. The largest Youden value was used as the cut-off point, and binomial logistic regression analysis was performed to analyze the PI and Emean . Receiver operating characteristic (ROC) curve analysis was performed to determine the diagnostic power of parameters, individually and in combination.

RESULTS

The PI, Emean , MVD, and CFC were significantly lower in the control group than in the experimental group (P < .05). The PI, Emean , MVD, and CFC were significantly higher for MIBC than for non-muscle-invasive bladder cancer (P < .05). There were significant correlations between PI and MVD, and between Emean and CFC. The diagnostic efficiency analysis showed PI had the highest sensitivity, CFC had the highest specificity, and PI + Emean had the highest diagnostic efficacy.

CONCLUSION

CEUS and elastography can distinguish lesions from normal tissue. PI, MVD, Emean , and CFC were useful for the detection of BLCA myometrial invasion. The comprehensive utilization of PI and Emean improved diagnostic accuracy and have clinical application.

Parametric maps of spatial two-tissue compartment model for prostate dynamic contrast enhanced MRI - comparison with the standard tofts model in the diagnosis of prostate cancer

The spatial two-tissue compartment model (2TCM) was used to analyze prostate dynamic contrast enhanced (DCE) MRI data and compared with the standard Tofts model. A total of 29 patients with biopsy-confirmed prostate cancer were included in this IRB-approved study. MRI data were acquired on a Philips Achieva 3T-TX scanner. After T2-weighted and diffusion-weighted imaging, DCE data using 3D T1-FFE mDIXON sequence were acquired pre- and post-contrast media injection (0.1 mmol/kg Multihance) for 60 dynamic scans with temporal resolution of 8.3 s/image. The 2TCM has one fast ([Formula: see text] and [Formula: see text]) and one slow ([Formula: see text] and [Formula: see text]) exchanging compartment, compared with the standard Tofts model parameters (Ktrans and kep). On average, prostate cancer had significantly higher values (p < 0.01) than normal prostate tissue for all calculated parameters. There was a strong correlation (r = 0.94, p < 0.001) between Ktrans and [Formula: see text] for cancer, but weak correlation (r = 0.28, p < 0.05) between kep and [Formula: see text]. Average root-mean-square error (RMSE) in fits from the 2TCM was significantly smaller (p < 0.001) than the RMSE in fits from the Tofts model. Receiver operating characteristic (ROC) analysis showed that fast [Formula: see text] had the highest area under the curve (AUC) than any other individual parameter. The combined four parameters from the 2TCM had a considerably higher AUC value than the combined two parameters from the Tofts model. The 2TCM is useful for quantitative analysis of prostate DCE-MRI data and provides new information in the diagnosis of prostate cancer.

Fine-Grained and Multiple Classification for Alzheimer's Disease With Wavelet Convolution Unit Network

In this article, we propose a novel wavelet convolution unit for the image-oriented neural network to integrate wavelet analysis with a vanilla convolution operator to extract deep abstract features more efficiently. On one hand, in order to acquire non-local receptive fields and avoid information loss, we define a new convolution operation by composing a traditional convolution function and approximate and detailed representations after single-scale wavelet decomposition of source images. On the other hand, multi-scale wavelet decomposition is introduced to obtain more comprehensive multi-scale feature information. Then, we fuse all these cross-scale features to improve the problem of inaccurate localization of singular points. Given the novel wavelet convolution unit, we further design a network based on it for fine-grained Alzheimer's disease classifications (i.e., Alzheimer's disease, Normal controls, early mild cognitive impairment, late mild cognitive impairment). Up to now, only a few methods have studied one or several fine-grained classifications, and even fewer methods can achieve both fine-grained and multi-class classifications. We adopt the novel network and diffuse tensor images to achieve fine-grained classifications, which achieved state-of-the-art accuracy for all eight kinds of fine-grained classifications, up to 97.30%, 95.78%, 95.00%, 94.00%, 97.89%, 95.71%, 95.07%, 93.79%. In order to build a reference standard for Alzheimer's disease classifications, we actually implemented all twelve coarse-grained and fine-grained classifications. The results show that the proposed method achieves solidly high accuracy for them. Its classification ability greatly exceeds any kind of existing Alzheimer's disease classification method.

Preoperative stricture length measurement does not predict postoperative outcomes in robotic ureteral reconstructive surgery

PURPOSE

We sought to determine whether preoperative stricture length measurement affected the choice of procedure performed, its correlation to intraoperative stricture length, and postoperative outcomes.

METHODS

The Collaborative of Reconstructive Robotic Ureteral Surgery (CORRUS) database was queried for patients undergoing robotic ureteral reconstructive surgery from 2013 to 2021 who had surgical stricture length measurement. From this cohort, we identified patients with and without preoperative stricture length measurement via retrograde pyelogram or antegrade nephrostogram. Outcomes evaluated included intraoperative complications, 30-day complications greater than Clavien-Dindo grade II, hardware-free status, and need for additional procedures.

RESULTS

Of 153 patients with surgical stricture length measurements, 102 (66.7%) had preoperative radiographic measurement. No repair type was more likely to have preoperative measurement. The Pearson correlation coefficient between surgical and radiographic stricture length measurements was + 0.79. The average surgical measurement was 0.71 cm (± 1.52) longer than radiographic assessment. Those with preoperative imaging waited on average 5.0 months longer for surgery, but this finding was not statistically significant (p = 0.18). There was no statistically significant difference in intraoperative complications, 30-day complication rates, hardware-free status at last follow-up, or need for additional procedures between patients with and without preoperative measurement. The only significant predictive factor was preoperative stricture length on 30-day postoperative complications.

CONCLUSIONS

Despite relatively high prevalence of preoperative radiographic stricture length measurement, there are few measures where it offers clinically meaningful diagnostic information towards the definitive surgical management of ureteral stricture disease.

Induction therapy with ipilimumab and nivolumab followed by consolidative chemoradiation as organ-sparing treatment in urothelial bladder cancer: study protocol of the INDIBLADE trial

Introduction

Studies that assessed the efficacy of pre-operative immune checkpoint blockade (ICB) in locally advanced urothelial cancer of the bladder showed encouraging pathological complete response rates, suggesting that a bladder-sparing approach may be a viable option in a subset of patients. Chemoradiation is an alternative for radical cystectomy with similar oncological outcomes, but is still mainly used in selected patients with organ-confined tumors or patients ineligible to undergo radical cystectomy. We propose to sequentially administer ICB and chemoradiation to patients with (locally advanced) muscle-invasive bladder cancer.

Methods

The INDIBLADE trial is an investigator-initiated, single-arm, multicenter phase 2 trial. Fifty patients with cT2-4aN0-2M0 urothelial bladder cancer will be treated with ipilimumab 3 mg/kg on day 1, ipilimumab 3 mg/kg plus nivolumab 1 mg/kg on day 22, and nivolumab 3 mg/kg on day 43 followed by chemoradiation. The primary endpoint is the bladder-intact event-free survival (BI-EFS). Events include: local or distant recurrence, salvage cystectomy, death and switch to platinum-based chemotherapy. We will also evaluate the potential of multiparametric magnetic resonance imaging of the bladder to identify non-responders, and we will assess the clearance of circulating tumor DNA as a biomarker for ICB treatment response.

Discussion

This is the first trial in which the efficacy of induction combination ICB followed by chemoradiation is being evaluated to provide bladder-preservation in patients with (locally advanced) urothelial bladder cancer.

Clinical Trial Registration

The INDIBLADE trial was registered on clinicaltrials.gov on January 21, 2022 (NCT05200988).

Deep Learning Methodologies Applied to Digital Pathology in Prostate Cancer: A Systematic Review

Deep learning (DL), often called artificial intelligence (AI), has been increasingly used in Pathology thanks to the use of scanners to digitize slides which allow us to visualize them on monitors and process them with AI algorithms. Many articles have focused on DL applied to prostate cancer (PCa). This systematic review explains the DL applications and their performances for PCa in digital pathology. Article research was performed using PubMed and Embase to collect relevant articles. A Risk of Bias (RoB) was assessed with an adaptation of the QUADAS-2 tool. Out of the 77 included studies, eight focused on pre-processing tasks such as quality assessment or staining normalization. Most articles (n = 53) focused on diagnosis tasks like cancer detection or Gleason grading. Fifteen articles focused on prediction tasks, such as recurrence prediction or genomic correlations. Best performances were reached for cancer detection with an Area Under the Curve (AUC) up to 0.99 with algorithms already available for routine diagnosis. A few biases outlined by the RoB analysis are often found in these articles, such as the lack of external validation. This review was registered on PROSPERO under CRD42023418661.

Pediatric renal trauma at a level 1 trauma center in a rural state: A 10-year institutional review and protocol implementation

INTRODUCTION

Unintentional injury is a leading cause of mortality and morbidity in children. There is no consensus on the ideal, discrete management of pediatric renal trauma (PRT). Therefore, management protocols tend to be institution-specific.

OBJECTIVE

This study aimed to characterize PRT at a rural level-1 trauma center and subsequently develop a standardized protocol.

STUDY DESIGN

A retrospective review of a prospectively maintained database of PRT at a rural level 1 trauma center between 2009 and 2019 was conducted. Injuries were characterized regarding renal trauma grade, associated multi-organ involvement and the need for intervention. The benefit of patient transfer from regional hospitals and length and cost of stay were evaluated.

RESULTS

Of 250 patients admitted with renal trauma diagnosis 50 patients <18 years were analyzed. Of those, the majority (32/50, 64%) had low-grade (grade I-III) injuries. Conservative management was successful in all low-grade injuries. Of 18 high-grade PRT, 10 (55.6%) required intervention, one prior to transfer. Among patients with low-grade trauma, 23/32 (72%) were transferred from an outside facility. A total of 13 (26%) patients with isolated low-grade renal trauma were transferred from regional hospitals. All isolated, transferred low-grade renal trauma had diagnostic imaging before transfer and none required invasive intervention. Interventional management of renal injury was associated with a longer median LOS [7 (IQR = 4-16.5) vs 4 (IQR = 2-6) days for conservative management, p = 0.019)] and an increased median total cost of $57,986 vs. $18,042 for conservative management (p = 0.002).

DISCUSSION

The majority of PRT, particularly low-grade, can be managed conservatively. A significant proportion of children with low-grade trauma are unnecessarily transferred to higher level centers. Review of pediatric renal trauma at our institution over a decade has allowed us to develop an institutional protocol which we believe allows for safe and effective patient monitoring.

CONCLUSION

Isolated, low-grade PRT can be managed conservatively at regional hospitals without needing transfer to a level 1 trauma center. Children with high-grade injuries should be closely monitored and are more likely to need invasive intervention. Development of a PRT protocol will help to safely triage this population and identify those who may benefit from transfer to a tertiary care center.

Estimating the Irreducible Uncertainty in Visual Diagnosis: Statistical Modeling of Skill Using Response Models

BACKGROUND

For the representative problem of prostate cancer grading, we sought to simultaneously model both the continuous nature of the case spectrum and the decision thresholds of individual pathologists, allowing quantitative comparison of how they handle cases at the borderline between diagnostic categories.

METHODS

Experts and pathology residents each rated a standardized set of prostate cancer histopathological images on the International Society of Urological Pathologists (ISUP) scale used in clinical practice. They diagnosed 50 histologic cases with a range of malignancy, including intermediate cases in which clear distinction was difficult. We report a statistical model showing the degree to which each individual participant can separate the cases along the latent decision spectrum.

RESULTS

The slides were rated by 36 physicians in total: 23 ISUP pathologists and 13 residents. As anticipated, the cases showed a full continuous range of diagnostic severity. Cases ranged along a logit scale consistent with the consensus rating (Consensus ISUP 1: mean -0.93 [95% confidence interval {CI} -1.10 to -0.78], ISUP 2: -0.19 logits [-0.27 to -0.12]; ISUP 3: 0.56 logits [0.06-1.06]; ISUP 4 1.24 logits [1.10-1.38]; ISUP 5: 1.92 [1.80-2.04]). The best raters were able to meaningfully discriminate between all 5 ISUP categories, showing intercategory thresholds that were quantifiably precise and meaningful.

CONCLUSIONS

We present a method that allows simultaneous quantification of both the confusability of a particular case and the skill with which raters can distinguish the cases.

IMPLICATIONS

The technique generalizes beyond the current example to other clinical situations in which a diagnostician must impose an ordinal rating on a biological spectrum.

HIGHLIGHTS

Question: How can we quantify skill in visual diagnosis for cases that sit at the border between 2 ordinal categories-cases that are inherently difficult to diagnose?Findings: In this analysis of pathologists and residents rating prostate biopsy specimens, decision-aligned response models are calculated that show how pathologists would be likely to classify any given case on the diagnostic spectrum. Decision thresholds are shown to vary in their location and precision.Significance: Improving on traditional measures such as kappa and receiver-operating characteristic curves, this specialization of item response models allows better individual feedback to both trainees and pathologists, including better quantification of acceptable decision variation.

A Brief Review of Artificial Intelligence in Genitourinary Oncological Imaging

Genitourinary (GU) system is among the most commonly involved malignancy sites in the human body. Imaging plays a crucial role not only in diagnosis of cancer but also in disease management and its prognosis. However, interpretation of conventional imaging methods such as CT or MR imaging (MRI) usually demonstrates variability across different readers and institutions. Artificial intelligence (AI) has emerged as a promising technology that could improve the patient care by providing helpful input to human readers through lesion detection algorithms and lesion classification systems. Moreover, the robustness of these models may be valuable in automating time-consuming tasks such as organ and lesion segmentations. Herein, we review the current state of imaging and existing challenges in GU malignancies, particularly for cancers of prostate, kidney and bladder; and briefly summarize the recent AI-based solutions to these challenges.

Immediately repaired penile fractures: age is the only predictor of postoperative long-term functional outcomes

Background

Penile fractures can lead to many functional complications, especially erectile dysfunction (ED). Few studies have evaluated the factors that predict late complications of an immediately repaired penile fracture.

Aim

To identify the potential predictors of long-term poor functional outcomes following immediate surgical intervention for penile fractures.

Methods

Sixty-eight consecutive patients with suspected penile fracture between 2003 and 2022 were retrospectively reviewed. Functional outcomes, postoperative complications, and follow-up duration were obtained from the records of follow-up visits. Age at presentation, location and length of the tunical tear, the presence of urethral rupture, and time to surgery were all analyzed as potential risk factors for postoperative functional outcomes.

Outcomes

Postoperative erectile function and intercourse satisfaction were measured by the IIEF-5 (the 5-item version of the International Index of Erectile Function). Penile curvature, a palpable nodule, and paresthesia/numbness were detected by physical examination. Uroflowmetry was used to assess urinary flow in patients who underwent urethral repair.

Results

Fifty-eight patients were analyzed. The mean ± SD age was 38.1 ± 10.4 years; the median follow-up was 79.0 months (range, 13-180); the median time to surgery was 9.8 hours (4-30); and the median tunical tear length was 15.5 mm (4-40). Urethral rupture was observed in 8 patients (13.8%). In univariable analyses, urethral rupture was associated with postoperative complications (P = .034). In addition, age at presentation and tunical tear size were significantly associated with postoperative complications and ED (P < .05). However, in multivariable analyses, only age at presentation significantly predicted postoperative complications and ED (P = .004 and P = .037).

Clinical Implications

Age at presentation is the most important factor determining the prognosis of immediate surgical repair of the penile fracture, which aids in predicting potential complications and discussing them with patients prior to surgical intervention and during the follow-up period.

Strengths and Limitations

The study's retrospective design is an important limitation. Furthermore, there were no data on an IIEF-5 outcome measuring preoperative erectile function.

Conclusion

These results revealed an association between (1) urethral rupture, longer tunical tears, and older age and (2) the development of late complications. The remarkable finding of this study was that age at presentation was the only significant predictor of functional complications based on multivariable analyses. This relationship also remained robust in tests evaluating the covariance of the effects of aging on ED.

Metadata and Image Features Co-Aware Personalized Federated Learning for Smart Healthcare

Recently, artificial intelligence has been widely used in intelligent disease diagnosis and has achieved great success. However, most of the works mainly rely on the extraction of image features but ignore the use of clinical text information of patients, which may limit the diagnosis accuracy fundamentally. In this paper, we propose a metadata and image features co-aware personalized federated learning scheme for smart healthcare. Specifically, we construct an intelligent diagnosis model, by which users can obtain fast and accurate diagnosis services. Meanwhile, a personalized federated learning scheme is designed to utilize the knowledge learned from other edge nodes with larger contributions and customize high-quality personalized classification models for each edge node. Subsequently, a Naïve Bayes classifier is devised for classifying patient metadata. And then the image and metadata diagnosis results are jointly aggregated by different weights to improve the accuracy of intelligent diagnosis. Finally, the simulation results illustrate that, compared with the existing methods, our proposed algorithm achieves better classification accuracy, reaching about 97.16% on PAD-UFES-20 dataset.

Magnetic Resonance Imaging in Clinical Trials of Diabetic Kidney Disease

Chronic kidney disease (CKD) associated with diabetes mellitus (DM) (known as diabetic kidney disease, DKD) is a serious and growing healthcare problem worldwide. In DM patients, DKD is generally diagnosed based on the presence of albuminuria and a reduced glomerular filtration rate. Diagnosis rarely includes an invasive kidney biopsy, although DKD has some characteristic histological features, and kidney fibrosis and nephron loss cause disease progression that eventually ends in kidney failure. Alternative sensitive and reliable non-invasive biomarkers are needed for DKD (and CKD in general) to improve timely diagnosis and aid disease monitoring without the need for a kidney biopsy. Such biomarkers may also serve as endpoints in clinical trials of new treatments. Non-invasive magnetic resonance imaging (MRI), particularly multiparametric MRI, may achieve these goals. In this article, we review emerging data on MRI techniques and their scientific, clinical, and economic value in DKD/CKD for diagnosis, assessment of disease pathogenesis and progression, and as potential biomarkers for clinical trial use that may also increase our understanding of the efficacy and mode(s) of action of potential DKD therapeutic interventions. We also consider how multi-site MRI studies are conducted and the challenges that should be addressed to increase wider application of MRI in DKD.

MRI-Based Radiomics in Bladder Cancer: A Systematic Review and Radiomics Quality Score Assessment

(1): Background: With the recent introduction of vesical imaging reporting and data system (VI-RADS), magnetic resonance imaging (MRI) has become the main imaging method used for the preoperative local staging of bladder cancer (BCa). However, the VI-RADS score is subject to interobserver variability and cannot provide information about tumor cellularity. These limitations may be overcome by using a quantitative approach, such as the new emerging domain of radiomics. (2) Aim: To systematically review published studies on the use of MRI-based radiomics in bladder cancer. (3) Materials and Methods: We performed literature research using the PubMed MEDLINE, Scopus, and Web of Science databases using PRISMA principles. A total of 1092 papers that addressed the use of radiomics for BC staging, grading, and treatment response were retrieved using the keywords "bladder cancer", "magnetic resonance imaging", "radiomics", and "textural analysis". (4) Results: 26 papers met the eligibility criteria and were included in the final review. The principal applications of radiomics were preoperative tumor staging (n = 13), preoperative prediction of tumor grade or molecular correlates (n = 9), and prediction of prognosis/response to neoadjuvant therapy (n = 4). Most of the developed radiomics models included second-order features mainly derived from filtered images. These models were validated in 16 studies. The average radiomics quality score was 11.7, ranging between 8.33% and 52.77%. (5) Conclusions: MRI-based radiomics holds promise as a quantitative imaging biomarker of BCa characterization and prognosis. However, there is still need for improving the standardization of image preprocessing, feature extraction, and external validation before applying radiomics models in the clinical setting.

Battle of the (Chat)Bots: Comparing Large Language Models to Practice Guidelines for Transfusion-Associated Graft-Versus-Host Disease Prevention

Published guidelines and clinical practices vary when defining indications for irradiation of blood components for the prevention of transfusion-associated graft-versus-host disease (TA-GVHD). This study assessed irradiation indication lists generated by multiple artificial intelligence (AI) programs, or chatbots, and compared them to 2020 British Society for Haematology (BSH) practice guidelines. Four chatbots (ChatGPT-3.5, ChatGPT-4, Bard, and Bing Chat) were prompted to list the indications for irradiation to prevent TA-GVHD. Responses were graded for concordance with BSH guidelines. Chatbot response length, discrepancies, and omissions were noted. Chatbot responses differed, but all were relevant, short in length, generally more concordant than discordant with BSH guidelines, and roughly complete. They lacked several indications listed in BSH guidelines and notably differed in their irradiation eligibility criteria for fetuses and neonates. The chatbots variably listed erroneous indications for TA-GVHD prevention, such as patients receiving blood from a donor who is of a different race or ethnicity. This study demonstrates the potential use of generative AI for transfusion medicine and hematology topics but underscores the risk of chatbot medical misinformation. Further study of risk factors for TA-GVHD, as well as the applications of chatbots in transfusion medicine and hematology, is warranted.

Improving clinical diagnostic accuracy and management of False penile fractures characterizing typical clinical presentation: a systematic review and meta-analysis

PURPOSE

False penile fractures (FPF) represent a rare sexual emergency characterized by blunt trauma of penis in the absence of albuginea's injury, with or without lesion of dorsal penile vein. Their presentation is often indistinguishable from true penile fractures (TPF). This overlapping of clinical presentation, and lack of knowledge about FPF, can lead surgeons often to proceed directly to surgical exploration without further examinations. The aim of this study was to define a typical presentation of false penile fractures (FPF) emergency, identifying in absence of "snap" sound, slow detumescence, penile shaft ecchymosis, and penile deviation main clinical signs.

METHODS

We performed a systematic review and meta-analysis based on Medline, Scopus and Cochrane following a protocol designed a priori, to define sensitivity of "snap" sound absence, slow detumescence and penile deviation.

RESULTS

Based on the literature search of 93 articles, 15 were included (73 patients). All patients referred pain, most of them during coitus (n = 57; 78%). Detumescence occurred in 37/73 (51%), and all patients described detumescence occurrence as "slow". The results show that single anamnestic item have a high-moderate sensibility in the diagnosis of FPF, and the highest sensitive item was penile deviation (sensibility = 0.86). However, when more than one item is present, overall sensitivity increases greatly, closing to 100% (95% Confidence Interval 92-100).

CONCLUSION

Surgeons can consciously decide between additional exams, a conservative approach, and rapid intervention using these indicators to detect FPF. Our findings identified symptoms with excellent specificity for FPF diagnosis, giving clinicians more useful tools for making decisions.

Identification of Origin for Spinal Metastases from MR Images: Comparison Between Radiomics and Deep Learning Methods

OBJECTIVE

To determine whether spinal metastatic lesions originated from lung cancer or from other cancers based on spinal contrast-enhanced T1 (CET1) magnetic resonance (MR) images analyzed using radiomics (RAD) and deep learning (DL) methods.

METHODS

We recruited and retrospectively reviewed 173 patients diagnosed with spinal metastases at two different centers between July 2018 and June 2021. Of these, 68 involved lung cancer and 105 were other types of cancer. They were assigned to an internal cohort of 149 patients, randomly divided into a training set and a validation set, and to an external cohort of 24 patients. All patients underwent CET1-MR imaging before surgery or biopsy. We developed two predictive algorithms: a DL model and a RAD model. We compared performance between models, and against human radiological assessment, via accuracy (ACC) and receiver operating characteristic (ROC) analyses. Furthermore, we analyzed the correlation between RAD and DL features.

RESULTS

The DL model outperformed RAD model across the board, with ACC/ area under the receiver operating characteristic curve (AUC) values of 0.93/0.94 (DL) versus 0.84/0.93 (RAD) when applied to the training set from the internal cohort, 0.74/0.76 versus 0.72/0.75 when applied to the validation set, and 0.72/0.76 versus 0.69/0.72 when applied to the external test cohort. For the validation set, it also outperformed expert radiological assessment (ACC: 0.65, AUC: 0.68). We only found weak correlations between DL and RAD features.

CONCLUSION

The DL algorithm successfully identified the origin of spinal metastases from pre-operative CET1-MR images, outperforming both RAD models and expert assessment by trained radiologists.

[Role of Interventional Radiologists in Trauma Centers]

Based on statistics available in Korea, trauma centers play a critical role in treatment of patients with trauma. Interventional radiologists in trauma centers perform various procedures, including embolization, which constitutes the basic treatment for control of hemorrhage, although interventions such as stent graft insertion may also be used. Although emergency interventional procedures have been used conventionally, rapid and effective hemorrhage control is important in patients with trauma. Therefore, it is important to accurately understand and implement the concept of damage control interventional radiology, which has gained attention in recent times, to reduce preventable trauma-induced mortality rates.

An Overview of Angiogenesis in Bladder Cancer

PURPOSE OF THE REVIEW

Angiogenesis plays a key role in bladder cancer (BC) pathogenesis. In the last two decades, an increasing number of publications depicting a multitude of novel angiogenic molecules and pathways have emerged. The growing complexity necessitates an evaluation of the breadth of current knowledge to highlight key findings and guide future research.

RECENT FINDINGS

Angiogenesis is a dynamic biologic process that is inherently difficult to assess. Clinical assessment of angiogenesis in BCs is advancing with the integration of image analysis systems and dynamic contrast-enhanced and magnetic resonance imaging (DCE-MRI). Tumour-associated macrophages (TAMs) significantly influence the angiogenic process, and further research is needed to assess their potential as therapeutic targets. A rapidly growing list of non-coding RNAs affect angiogenesis in BCs, partly through modulation of vascular endothelial growth factor (VEGF) activity. Vascular mimicry (VM) has been repeatedly associated with increased tumour aggressiveness in BCs. Standardised assays are needed for appropriate identification and quantification of VM channels. This article demonstrates the dynamic and complex nature of the angiogenic process and asserts the need for further studies to deepen our understanding.

Multi-view radiomics and deep learning modeling for prostate cancer detection based on multi-parametric MRI

Introduction

This study aims to develop an imaging model based on multi-parametric MR images for distinguishing between prostate cancer (PCa) and prostate hyperplasia.

Methods

A total of 236 subjects were enrolled and divided into training and test sets for model construction. Firstly, a multi-view radiomics modeling strategy was designed in which different combinations of radiomics feature categories (original, LoG, and wavelet) were compared to obtain the optimal input feature sets. Minimum-redundancy maximum-relevance (mRMR) selection and least absolute shrinkage selection operator (LASSO) were used for feature reduction, and the next logistic regression method was used for model construction. Then, a Swin Transformer architecture was designed and trained using transfer learning techniques to construct the deep learning models (DL). Finally, the constructed multi-view radiomics and DL models were combined and compared for model selection and nomogram construction. The prediction accuracy, consistency, and clinical benefit were comprehensively evaluated in the model comparison.

Results

The optimal input feature set was found when LoG and wavelet features were combined, while 22 and 17 radiomic features in this set were selected to construct the ADC and T2 multi-view radiomic models, respectively. ADC and T2 DL models were built by transferring learning from a large number of natural images to a relatively small sample of prostate images. All individual and combined models showed good predictive accuracy, consistency, and clinical benefit. Compared with using only an ADC-based model, adding a T2-based model to the combined model would reduce the model's predictive performance. The ADCCombinedScore model showed the best predictive performance among all and was transformed into a nomogram for better use in clinics.

Discussion

The constructed models in our study can be used as a predictor in differentiating PCa and BPH, thus helping clinicians make better clinical treatment decisions and reducing unnecessary prostate biopsies.

Kidney Segmentation from Dynamic Contrast-Enhanced Magnetic Resonance Imaging Integrating Deep Convolutional Neural Networks and Level Set Methods

The dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) technique has taken on a significant and increasing role in diagnostic procedures and treatments for patients who suffer from chronic kidney disease. Careful segmentation of kidneys from DCE-MRI scans is an essential early step towards the evaluation of kidney function. Recently, deep convolutional neural networks have increased in popularity in medical image segmentation. To this end, in this paper, we propose a new and fully automated two-phase approach that integrates convolutional neural networks and level set methods to delimit kidneys in DCE-MRI scans. We first develop two convolutional neural networks that rely on the U-Net structure (UNT) to predict a kidney probability map for DCE-MRI scans. Then, to leverage the segmentation performance, the pixel-wise kidney probability map predicted from the deep model is exploited with the shape prior information in a level set method to guide the contour evolution towards the target kidney. Real DCE-MRI datasets of 45 subjects are used for training, validating, and testing the proposed approach. The valuation results demonstrate the high performance of the two-phase approach, achieving a Dice similarity coefficient of 0.95 ± 0.02 and intersection over union of 0.91 ± 0.03, and 1.54 ± 1.6 considering a 95% Hausdorff distance. Our intensive experiments confirm the potential and effectiveness of that approach over both UNT models and numerous recent level set-based methods.

Optimal length and temporal resolution of dynamic contrast-enhanced MR imaging for the differentiation between prostate cancer and normal peripheral zone tissue

The value of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in the detection of prostate cancer is controversial. There are currently insufficient peer reviewed published data or expert consensus to support routine adoption of DCE-MRI for clinical use. Thus, the objective of this study was to explore the optimal temporal resolution and measurement length for DCE-MRI to differentiate cancerous from normal prostate tissue of the peripheral zone of the prostate by non-parametric MRI analysis and to compare with a quantitative MRI analysis. Predictors of interest were onset time, relative signal intensity (RSI), wash-in slope, peak enhancement, wash-out and wash-out slope determined from non-parametric characterisation of DCE-MRI intensity-time profiles. The discriminatory power was estimated from C-statistics based on cross validation. We analyzed 54 patients with 97 prostate tissue specimens (47 prostate cancer, 50 normal prostate tissue) of the peripheral zone, mean age 63.8 years, mean prostate-specific antigen 18.9 ng/mL and mean of 10.5 days between MRI and total prostatectomy. When comparing prostate cancer tissue with normal prostate tissue, median RSI was 422% vs 330%, and wash-in slope 0.870 vs 0.539. The peak enhancement of 67 vs 42 was higher with prostate cancer tissue, while wash-out (-30% vs -23%) and wash-out slope (-0.037 vs -0.029) were lower, and the onset time (32 seconds) was comparable. The optimal C-statistics was 0.743 for temporal resolution of 8.0 seconds and measurement length of 2.5 minutes compared with 0.656 derived from a quantitative MRI analysis. This study provides evidence that the use of a non-parametric approach instead of a more established parametric approach resulted in greater precision to differentiate cancerous from normal prostate tissue of the peripheral zone of the prostate.

Expanding Horizons: The Realities of CAD, the Promise of Artificial Intelligence, and Machine Learning's Role in Breast Imaging beyond Screening Mammography

Artificial intelligence (AI) applications in mammography have gained significant popular attention; however, AI has the potential to revolutionize other aspects of breast imaging beyond simple lesion detection. AI has the potential to enhance risk assessment by combining conventional factors with imaging and improve lesion detection through a comparison with prior studies and considerations of symmetry. It also holds promise in ultrasound analysis and automated whole breast ultrasound, areas marked by unique challenges. AI's potential utility also extends to administrative tasks such as MQSA compliance, scheduling, and protocoling, which can reduce the radiologists' workload. However, adoption in breast imaging faces limitations in terms of data quality and standardization, generalizability, benchmarking performance, and integration into clinical workflows. Developing methods for radiologists to interpret AI decisions, and understanding patient perspectives to build trust in AI results, will be key future endeavors, with the ultimate aim of fostering more efficient radiology practices and better patient care.

Evaluation of Whole-Tumor Texture Analysis Based on MRI Diffusion Kurtosis and Biparametric VI-RADS Model for Staging and Grading Bladder Cancer

BACKGROUND

to evaluate the feasibility of texture analysis (TA) based on diffusion kurtosis imaging (DKI) in staging and grading bladder cancer (BC) and to compare it with apparent diffusion coefficient (ADC) and biparametric vesical imaging reporting and data system (VI-RADS).

MATERIALS AND METHODS

In this retrospective study, 101 patients with pathologically confirmed BC underwent MRI with multiple-b values ranging from 0 to 2000 s/mm². ADC- and DKI-derived parameters, including mean kurtosis (MK) and mean diffusivity (MD), were obtained. First-order texture histogram parameters of MK and MD, including the mean; 5th, 25th, 50th, 75th, and 90th percentiles; inhomogeneity; skewness: kurtosis; and entropy; were extracted. The VI-RADS score was evaluated based on the T2WI and DWI. The Mann-Whitney U-test was used to compare the texture parameters and ADC values between non-muscle-invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC), as well as between low and high grades. Receiver operating characteristic analysis was used to evaluate the diagnostic performance of each significant parameter and their combinations.

RESULTS

The NMIBC and low-grade group had higher MDmean, MD5th, MD25th, MD50th, MD75th, MD90th, and ADC values than those of the MIBC and the high-grade group. The NMIBC and low-grade group yielded lower MKmean, MK25th, MK50th, MK75th, and MK90th than the MIBC and high-grade group. Among all histogram parameters, MD75th and MD90th yielded the highest AUC in differentiating MIBC from NMIBC (both AUCs were 0.87), while the AUC for ADC was 0.86. The MK75th and MK90th had the highest AUC (both 0.79) in differentiating low- from high-grade BC, while ADC had an AUC of 0.68. The AUC (0.92) of the combination of DKI histogram parameters (MD75th, MD90th, and MK90th) with biparametric VI-RADS in staging BC was higher than that of the biparametric VI-RADS (0.89).

CONCLUSIONS

Texture-analysis-derived DKI is useful in evaluating both the staging and grading of bladder cancer; in addition, the histogram parameters of the DKI (MD75th, MD90th, and MK90th) can provide additional value to VI-RADS.

Anatomically guided self-adapting deep neural network for clinically significant prostate cancer detection on bi-parametric MRI: a multi-center study

OBJECTIVE

To evaluate the effectiveness of a self-adapting deep network, trained on large-scale bi-parametric MRI data, in detecting clinically significant prostate cancer (csPCa) in external multi-center data from men of diverse demographics; to investigate the advantages of transfer learning.

METHODS

We used two samples: (i) Publicly available multi-center and multi-vendor Prostate Imaging: Cancer AI (PI-CAI) training data, consisting of 1500 bi-parametric MRI scans, along with its unseen validation and testing samples; (ii) In-house multi-center testing and transfer learning data, comprising 1036 and 200 bi-parametric MRI scans. We trained a self-adapting 3D nnU-Net model using probabilistic prostate masks on the PI-CAI data and evaluated its performance on the hidden validation and testing samples and the in-house data with and without transfer learning. We used the area under the receiver operating characteristic (AUROC) curve to evaluate patient-level performance in detecting csPCa.

RESULTS

The PI-CAI training data had 425 scans with csPCa, while the in-house testing and fine-tuning data had 288 and 50 scans with csPCa, respectively. The nnU-Net model achieved an AUROC of 0.888 and 0.889 on the hidden validation and testing data. The model performed with an AUROC of 0.886 on the in-house testing data, with a slight decrease in performance to 0.870 using transfer learning.

CONCLUSIONS

The state-of-the-art deep learning method using prostate masks trained on large-scale bi-parametric MRI data provides high performance in detecting csPCa in internal and external testing data with different characteristics, demonstrating the robustness and generalizability of deep learning within and across datasets.

CLINICAL RELEVANCE STATEMENT

A self-adapting deep network, utilizing prostate masks and trained on large-scale bi-parametric MRI data, is effective in accurately detecting clinically significant prostate cancer across diverse datasets, highlighting the potential of deep learning methods for improving prostate cancer detection in clinical practice.

MSF-Net: A Lightweight Multi-Scale Feature Fusion Network for Skin Lesion Segmentation

Segmentation of skin lesion images facilitates the early diagnosis of melanoma. However, this remains a challenging task due to the diversity of target scales, irregular segmentation shapes, low contrast, and blurred boundaries of dermatological graphics. This paper proposes a multi-scale feature fusion network (MSF-Net) based on comprehensive attention convolutional neural network (CA-Net). We introduce the spatial attention mechanism in the convolution block through the residual connection to focus on the key regions. Meanwhile, Multi-scale Dilated Convolution Modules (MDC) and Multi-scale Feature Fusion Modules (MFF) are introduced to extract context information across scales and adaptively adjust the receptive field size of the feature map. We conducted many experiments on the public data set ISIC2018 to verify the validity of MSF-Net. The ablation experiment demonstrated the effectiveness of our three modules. The comparison experiment with the existing advanced network confirms that MSF-Net can achieve better segmentation under fewer parameters.

Dynamic contrast enhanced (DCE) MRI estimation of vascular parameters using knowledge-based adaptive models

We introduce and validate four adaptive models (AMs) to perform a physiologically based Nested-Model-Selection (NMS) estimation of such microvascular parameters as forward volumetric transfer constant, Ktrans, plasma volume fraction, vp, and extravascular, extracellular space, ve, directly from Dynamic Contrast-Enhanced (DCE) MRI raw information without the need for an Arterial-Input Function (AIF). In sixty-six immune-compromised-RNU rats implanted with human U-251 cancer cells, DCE-MRI studies estimated pharmacokinetic (PK) parameters using a group-averaged radiological AIF and an extended Patlak-based NMS paradigm. One-hundred-ninety features extracted from raw DCE-MRI information were used to construct and validate (nested-cross-validation, NCV) four AMs for estimation of model-based regions and their three PK parameters. An NMS-based a priori knowledge was used to fine-tune the AMs to improve their performance. Compared to the conventional analysis, AMs produced stable maps of vascular parameters and nested-model regions less impacted by AIF-dispersion. The performance (Correlation coefficient and Adjusted R-squared for NCV test cohorts) of the AMs were: 0.914/0.834, 0.825/0.720, 0.938/0.880, and 0.890/0.792 for predictions of nested model regions, vp, Ktrans, and ve, respectively. This study demonstrates an application of AMs that quickens and improves DCE-MRI based quantification of microvasculature properties of tumors and normal tissues relative to conventional approaches.

All That Glitters in cfDNA Analysis Is Not Gold or Its Utility Is Completely Established Due to Graft Damage: A Critical Review in the Field of Transplantation

In kidney transplantation, a biopsy is currently the gold standard for monitoring the transplanted organ. However, this is far from an ideal screening method given its invasive nature and the discomfort it can cause the patient. Large-scale studies in renal transplantation show that approximately 1% of biopsies generate major complications, with a risk of macroscopic hematuria greater than 3.5%. It would not be until 2011 that a method to detect donor-derived cell-free DNA (dd-cfDNA) employing digital PCR was devised based on analyzing the differences in SNPs between the donor and recipient. In addition, since the initial validation studies were carried out at the specific moments in which rejection was suspected, there is still not a good understanding of how dd-cfDNA levels naturally evolve post-transplant. In addition, various factors, both in the recipient and the donor, can influence dd-cfDNA levels and cause increases in the levels of dd-cfDNA themselves without suspicion of rejection. All that glitters in this technology is not gold; therefore, in this article, we discuss the current state of clinical studies, the benefits, and disadvantages.

Quantitative dynamic contrast-enhanced parameters and intravoxel incoherent motion facilitate the prediction of TP53 status and risk stratification of early-stage endometrial carcinoma

BACKGROUND

The aim of the study was to investigate the value of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and intravoxel incoherent motion (IVIM) in differentiating TP53-mutant from wild type, low-risk from non-low-risk early-stage endometrial carcinoma (EC).

PATIENTS AND METHODS

A total of 74 EC patients underwent pelvic MRI. Parameters volume transfer constant (K^trans), rate transfer constant (K_ep), the volume of extravascular extracellular space per unit volume of tissue (V_e), true diffusion coefficient (D), pseudo-diffusion coefficient (D*), and microvascular volume fraction (f) were compared. The combination of parameters was investigated by logistic regression and evaluated by bootstrap (1000 samples), receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).

RESULTS

In the TP53-mutant group, K^trans and K_ep were higher and D was lower than in the TP53-wild group; K^trans, V_e, f, and D were lower in the non-low-risk group than in the low-risk group (all P < 0.05). In the identification of TP53-mutant and TP53-wild early-stage EC, K^trans and D were independent predictors, and the combination of them had an optimal diagnostic efficacy (AUC, 0.867; sensitivity, 92.00%; specificity, 80.95%), which was significantly better than D (Z = 2.169, P = 0.030) and K^trans (Z = 2.572, P = 0.010). In the identification of low-risk and non-low-risk early-stage EC, K^trans, V_e, and f were independent predictors, and the combination of them had an optimal diagnostic efficacy (AUC, 0.947; sensitivity, 83.33%; specificity, 93.18%), which was significantly better than D (Z = 3.113, P = 0.002), f (Z = 4.317, P < 0.001), K^trans (Z = 2.713, P = 0.007), and V_e (Z = 3.175, P = 0.002). The calibration curves showed that the above two combinations of independent predictors, both have good consistency, and DCA showed that these combinations were reliable clinical prediction tools.

CONCLUSIONS

Both DCE-MRI and IVIM facilitate the prediction of TP53 status and risk stratification in early-stage EC. Compare with each single parameter, the combination of independent predictors provided better predictive power and may serve as a superior imaging marker.

Tissue response curve-shape analysis of dynamic glucose-enhanced and dynamic contrast-enhanced magnetic resonance imaging in patients with brain tumor

Dynamic glucose-enhanced (DGE) MRI is used to study the signal intensity time course (tissue response curve) after D-glucose injection. D-glucose has potential as a biodegradable alternative or complement to gadolinium-based contrast agents, with DGE being comparable with dynamic contrast-enhanced (DCE) MRI. However, the tissue uptake kinetics as well as the detection methods of DGE differ from DCE MRI, and it is relevant to compare these techniques in terms of spatiotemporal enhancement patterns. This study aims to develop a DGE analysis method based on tissue response curve shapes, and to investigate whether DGE MRI provides similar or complementary information to DCE MRI. Eleven patients with suspected gliomas were studied. Tissue response curves were measured for DGE and DCE MRI at 7 T and the area under the curve (AUC) was assessed. Seven types of response curve shapes were postulated and subsequently identified by deep learning to create color-coded "curve maps" showing the spatial distribution of different curve types. DGE AUC values were significantly higher in lesions than in normal tissue (p < 0.007). Furthermore, the distribution of curve types differed between lesions and normal tissue for both DGE and DCE. The DGE and DCE response curves in a 6-min postinjection time interval were classified as the same curve type in 20% of the lesion voxels, which increased to 29% when a 12-min DGE time interval was considered. While both DGE and DCE tissue response curve-shape analysis enabled differentiation of lesions from normal brain tissue in humans, their enhancements were neither temporally identical nor confined entirely to the same regions. Curve maps can provide accessible and intuitive information about the shape of DGE response curves, which is expected to be useful in the continued work towards the interpretation of DGE uptake curves in terms of D-glucose delivery, transport, and metabolism.

Advanced MR Techniques for Preoperative Glioma Characterization: Part 1

Preoperative clinical magnetic resonance imaging (MRI) protocols for gliomas, brain tumors with dismal outcomes due to their infiltrative properties, still rely on conventional structural MRI, which does not deliver information on tumor genotype and is limited in the delineation of diffuse gliomas. The GliMR COST action wants to raise awareness about the state of the art of advanced MRI techniques in gliomas and their possible clinical translation or lack thereof. This review describes current methods, limits, and applications of advanced MRI for the preoperative assessment of glioma, summarizing the level of clinical validation of different techniques. In this first part, we discuss dynamic susceptibility contrast and dynamic contrast-enhanced MRI, arterial spin labeling, diffusion-weighted MRI, vessel imaging, and magnetic resonance fingerprinting. The second part of this review addresses magnetic resonance spectroscopy, chemical exchange saturation transfer, susceptibility-weighted imaging, MRI-PET, MR elastography, and MR-based radiomics applications. Evidence Level: 3 Technical Efficacy: Stage 2.

A2OURSR: Adaptive adjustment based real MRI super-resolution via opinion-unaware measurements

High-quality and high-resolution magnetic resonance (MR) images can provide more details for diagnosis and analyses. Recently, MR images guided neurosurgery has become an emerging technique in clinics. Unlike other medical imaging techniques, it is impossible to achieve both real-time imaging and high image quality in MR imaging. The real-time performance is closely related to the nuclear magnetic equipment itself as well as the collection strategy of the k space data. Optimizing the imaging time cost via the corresponding algorithm is harder than enhancing image quality. Further, in reconstructing low-resolution and noise-rich MR images, getting relatively high-definition and resolution MR images as references are difficult or impossible. In addition, the existing methods are restricted in learning the controllable functions under the supervision of known degradation types and levels. As a result, severely bad results are inevitable when the modeling assumptions are far apart from the actual situation. To address these problems, we propose a novel adaptive adjustment method based on real MR images via opinion-unaware measurements for real super-resolution (A²OURSR). It can estimate the degree of blur and noise from the test image itself using two scores. These two scores can be considered pseudo labels to train the adaptive adjustable degradation estimation module. Then, the outputs of the above model are used as the inputs of the conditional network to tweak the generated results. Thus, the results can be automatically adjusted via the whole dynamic model. Extensive experimental results show that the proposed A²OURSR is superior to state-of-the-art methods on benchmarks quantitatively and visually.

A convolutional neural network-based model that predicts acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation

BACKGROUND

Forecasting acute graft-versus-host disease (aGVHD) after allogeneic hematopoietic stem cell transplantation (HSCT) is highly challenging with conventional statistical techniques due to complex parameters and their interactions. The primary object of this study was to establish a convolutional neural network (CNN)-based prediction model for aGVHD.

METHOD

We analyzed adult patients who underwent allogeneic HSCT between 2008 and 2018, using the Japanese nationwide registry database. The CNN algorithm, equipped with a natural language processing technique and an interpretable explanation algorithm, was applied to develop and validate prediction models.

RESULTS

Here, we evaluate 18,763 patients between 16 and 80 years of age (median, 50 years). In total, grade II-IV and grade III-IV aGVHD is observed among 42.0% and 15.6%. The CNN-based model eventually allows us to calculate a prediction score of aGVHD for an individual case, which is validated to distinguish the high-risk group of aGVHD in the test cohort: cumulative incidence of grade III-IV aGVHD at Day 100 after HSCT is 28.8% for patients assigned to a high-risk group by the CNN model, compared to 8.4% among low-risk patients (hazard ratio, 4.02; 95% confidence interval, 2.70-5.97; p < 0.01), suggesting high generalizability. Furthermore, our CNN-based model succeeds in visualizing the learning process. Moreover, contributions of pre-transplant parameters other than HLA information to the risk of aGVHD are determined.

CONCLUSIONS

Our results suggest that CNN-based prediction provides a faithful prediction model for aGVHD, and can serve as a valuable tool for decision-making in clinical practice.

Impact of Arterial Input Function and Pharmacokinetic Models on DCE-MRI Biomarkers for Detection of Vascular Effect Induced by Stroma-Directed Drug in an Orthotopic Mouse Model of Pancreatic Cancer

PURPOSE

We demonstrated earlier in mouse models of pancreatic ductal adenocarcinoma (PDA) that K^trans derived from dynamic contrast-enhanced (DCE) MRI detected microvascular effect induced by PEGPH20, a hyaluronidase which removes stromal hyaluronan, leading to reduced interstitial fluid pressure in the tumor (Clinical Cancer Res (2019) 25: 2314-2322). How the choice of pharmacokinetic (PK) model and arterial input function (AIF) may impact DCE-derived markers for detecting such an effect is not known.

PROCEDURES

Retrospective analyses of the DCE-MRI of the orthotopic PDA model are performed to examine the impact of individual versus group AIF combined with Tofts model (TM), extended-Tofts model (ETM), or shutter-speed model (SSM) on the ability to detect the microvascular changes induced by PEGPH20 treatment.

RESULTS

Individual AIF exhibit a marked difference in peak gadolinium concentration. However, across all three PK models, k_ep values show a significant correlation between individual versus group-AIF (p < 0.01). Regardless individual or group AIF, when k_ep is obtained from fitting the DCE-MRI data using the SSM, k_ep shows a significant increase after PEGPH20 treatment (p < 0.05 compared to the baseline); %change of k_ep from baseline to post-treatment is also significantly different between PEGPH20 versus vehicle group (p < 0.05). In comparison, when k_ep is derived from the TM, only the use of individual AIF leads to a significant increase of k_ep after PEGPH20 treatment, whereas the %change of k_ep is not different between PEGPH20 versus vehicle group. Group AIF but not individual AIF allows detection of a significant increase of V_p (derived from the ETM) in PEGPH20 versus vehicle group (p < 0.05). Increase of V_p is consistent with a large increase of mean capillary lumen area estimated from immunostaining.

CONCLUSION

Our results suggest that k_ep derived from SSM and V_p from ETM, both using group AIF, are optimal for the detection of microvascular changes induced by stroma-directed drug PEGPH20. These analyses provide insights in the choice of PK model and AIF for optimal DCE protocol design in mouse pancreatic cancer models.

Noninvasive Estimation of Tumor Interstitial Fluid Pressure from Subharmonic Scattering of Ultrasound Contrast Microbubbles

The noninvasive estimation of interstitial fluid pressure (IFP) using ultrasound contrast agent (UCA) microbubbles as pressure sensors will provide tumor treatments and efficacy assessments with a promising tool. This study aimed to verify the efficacy of the optimal acoustic pressure in vitro in the prediction of tumor IFPs based on UCA microbubbles' subharmonic scattering. A customized ultrasound scanner was used to generate subharmonic signals from microbubbles' nonlinear oscillations, and the optimal acoustic pressure was determined in vitro when the subharmonic amplitude reached the most sensitive to hydrostatic pressure changes. This optimal acoustic pressure was then applied to predict IFPs in tumor-bearing mouse models, which were further compared with the reference IFPs measured using a standard tissue fluid pressure monitor. An inverse linear relationship and good correlation (r = -0.853, p < 0.001) existed between the subharmonic amplitude and tumor IFPs at the optimal acoustic pressure of 555 kPa, and pressure sensitivity was 1.019 dB/mmHg. No statistical differences were found between the pressures measured by the standard device and those estimated via the subharmonic amplitude, as confirmed by cross-validation (mean absolute errors from 2.00 to 3.09 mmHg, p > 0.05). Our findings demonstrated that in vitro optimized acoustic parameters for UCA microbubbles' subharmonic scattering can be applied for the noninvasive estimation of tumor IFPs.

Predicting muscle invasion in bladder cancer based on MRI: A comparison of radiomics, and single-task and multi-task deep learning

BACKGROUND AND OBJECTIVES

Radiomics and deep learning are two popular technologies used to develop computer-aided detection and diagnosis schemes for analysing medical images. This study aimed to compare the effectiveness of radiomics, single-task deep learning (DL) and multi-task DL methods in predicting muscle-invasive bladder cancer (MIBC) status based on T2-weighted imaging (T2WI).

METHODS

A total of 121 tumours (93 for training, from Centre 1; 28 for testing, from Centre 2) were included. MIBC was confirmed with pathological examination. A radiomics model, a single-task model, and a multi-task model based on T2WI were constructed in the training cohort with five-fold cross-validation, and validation was conducted in the external test cohort. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of each model. DeLong's test and a permutation test were used to compare the performance of the models.

RESULTS

The area under the ROC curve (AUC) values of the radiomics, single-task and multi-task models in the training cohort were: 0.920, 0.933 and 0.932, respectively; and were 0.844, 0.884 and 0.932, respectively, in the test cohort. The multi-task model achieved better performance in the test cohort than did the other models. No statistically significant differences in AUC values and Kappa coefficients were observed between pairwise models, in either the training or test cohorts. According to the Grad-CAM feature visualization results, the multi-task model focused more on the diseased tissue area in some samples of the test cohort compared with the single-task model.

CONCLUSIONS

The T2WI-based radiomics, single-task, and multi-task models all exhibited good diagnostic performance in preoperatively predicting MIBC, in which the multi-task model had the best diagnostic performance. Compared with the radiomics method, our multi-task DL method had the advantage of saving time and effort. Compared with the single-task DL method, our multi-task DL method had the advantage of being more lesion-focused and more reliable for clinical reference.

PET in bladder cancer imaging

PURPOSE OF REVIEW

The aim of this study was to summarize the available evidence on different PET imaging modalities for the staging of patients diagnosed with bladder cancer (BCa). We further discuss the use of PET/computed tomography (CT) and PET/MRI with different radiopharmaceuticals to characterize tumour biology for treatment guidance.

RECENT FINDINGS

Available evidence supports the benefits of PET/CT in BCa staging due to its higher accuracy in the detection of nodal metastases compared with CT alone. The use of PET/MRI is of major future interest due to the higher soft tissue contrast of MRI, which might enable the early detection of the tumour in the bladder. For the time being, the sensitivity of PET/MRI is still too low, when it comes to the diagnosis of early-stage BCa. This is mainly due to the renal excretion of the commonly used [ 18 F]FDG PET tracer, wherefore small lesions in the wall of the bladder can be missed. Novel studies using PET radiopharmaceuticals to target immune checkpoints or other immune cell targets (immunoPET) demonstrated high uptake in tumour lesions with high PD-L1 expression. The use of immunoPET could therefore help identify BCa patients who exhibit PD-L1 positive tumours for systemic immune-therapy.

SUMMARY

PET/CT and PET/MRI seem to be promising imaging tools in BCa staging, especially for the detection of lymph node and distant metastases, as they are more accurate than conventional CT. Future clinical trials with novel radiopharmaceuticals and machine-learning driven PET-technologies bear the potential to help in the early detection, staging, monitoring and precision-medicine approach. Specifically, immunoPET is of high future interest, as it could help develop the concept of precision-medicine in the age of immunotherapy.

Recent Advances in Explainable Artificial Intelligence for Magnetic Resonance Imaging

Advances in artificial intelligence (AI), especially deep learning (DL), have facilitated magnetic resonance imaging (MRI) data analysis, enabling AI-assisted medical image diagnoses and prognoses. However, most of the DL models are considered as "black boxes". There is an unmet need to demystify DL models so domain experts can trust these high-performance DL models. This has resulted in a sub-domain of AI research called explainable artificial intelligence (XAI). In the last decade, many experts have dedicated their efforts to developing novel XAI methods that are competent at visualizing and explaining the logic behind data-driven DL models. However, XAI techniques are still in their infancy for medical MRI image analysis. This study aims to outline the XAI applications that are able to interpret DL models for MRI data analysis. We first introduce several common MRI data modalities. Then, a brief history of DL models is discussed. Next, we highlight XAI frameworks and elaborate on the principles of multiple popular XAI methods. Moreover, studies on XAI applications in MRI image analysis are reviewed across the tissues/organs of the human body. A quantitative analysis is conducted to reveal the insights of MRI researchers on these XAI techniques. Finally, evaluations of XAI methods are discussed. This survey presents recent advances in the XAI domain for explaining the DL models that have been utilized in MRI applications.

AV, Mahale A. Multimodal medical tensor fusion network-based DL framework for abnormality prediction from the radiology CXRs and clinical text reports

Pulmonary disease is a commonly occurring abnormality throughout this world. The pulmonary diseases include Tuberculosis, Pneumothorax, Cardiomegaly, Pulmonary atelectasis, Pneumonia, etc. A timely prognosis of pulmonary disease is essential. Increasing progress in Deep Learning (DL) techniques has significantly impacted and contributed to the medical domain, specifically in leveraging medical imaging for analysis, prognosis, and therapeutic decisions for clinicians. Many contemporary DL strategies for radiology focus on a single modality of data utilizing imaging features without considering the clinical context that provides more valuable complementary information for clinically consistent prognostic decisions. Also, the selection of the best data fusion strategy is crucial when performing Machine Learning (ML) or DL operation on multimodal heterogeneous data. We investigated multimodal medical fusion strategies leveraging DL techniques to predict pulmonary abnormality from the heterogeneous radiology Chest X-Rays (CXRs) and clinical text reports. In this research, we have proposed two effective unimodal and multimodal subnetworks to predict pulmonary abnormality from the CXR and clinical reports. We have conducted a comprehensive analysis and compared the performance of unimodal and multimodal models. The proposed models were applied to standard augmented data and the synthetic data generated to check the model's ability to predict from the new and unseen data. The proposed models were thoroughly assessed and examined against the publicly available Indiana university dataset and the data collected from the private medical hospital. The proposed multimodal models have given superior results compared to the unimodal models.

A two-stage renal disease classification based on transfer learning with hyperparameters optimization

Renal diseases are common health problems that affect millions of people around the world. Among these diseases, kidney stones, which affect anywhere from 1 to 15% of the global population and thus; considered one of the leading causes of chronic kidney diseases (CKD). In addition to kidney stones, renal cancer is the tenth most prevalent type of cancer, accounting for 2.5% of all cancers. Artificial intelligence (AI) in medical systems can assist radiologists and other healthcare professionals in diagnosing different renal diseases (RD) with high reliability. This study proposes an AI-based transfer learning framework to detect RD at an early stage. The framework presented on CT scans and images from microscopic histopathological examinations will help automatically and accurately classify patients with RD using convolutional neural network (CNN), pre-trained models, and an optimization algorithm on images. This study used the pre-trained CNN models VGG16, VGG19, Xception, DenseNet201, MobileNet, MobileNetV2, MobileNetV3Large, and NASNetMobile. In addition, the Sparrow search algorithm (SpaSA) is used to enhance the pre-trained model's performance using the best configuration. Two datasets were used, the first dataset are four classes: cyst, normal, stone, and tumor. In case of the latter, there are five categories within the second dataset that relate to the severity of the tumor: Grade 0, Grade 1, Grade 2, Grade 3, and Grade 4. DenseNet201 and MobileNet pre-trained models are the best for the four-classes dataset compared to others. Besides, the SGD Nesterov parameters optimizer is recommended by three models, while two models only recommend AdaGrad and AdaMax. Among the pre-trained models for the five-class dataset, DenseNet201 and Xception are the best. Experimental results prove the superiority of the proposed framework over other state-of-the-art classification models. The proposed framework records an accuracy of 99.98% (four classes) and 100% (five classes).

Rectal Culture-Based Versus Empirical Antibiotic Prophylaxis to Prevent Infectious Complications in Men Undergoing Transrectal Prostate Biopsy: A Randomized, Nonblinded Multicenter Trial

BACKGROUND

An increase in infections after transrectal prostate biopsy (PB), related to an increasing number of patients with ciprofloxacin-resistant rectal flora, necessitates the exploration of alternatives for the traditionally used empirical prophylaxis of ciprofloxacin. We compared infectious complication rates after transrectal PB using empirical ciprofloxacin prophylaxis versus culture-based prophylaxis.

METHODS

In this nonblinded, randomized trial, between 4 April 2018 and 30 July 2021, we enrolled 1538 patients from 11 Dutch hospitals undergoing transrectal PB. After rectal swab collection, patients were randomized 1:1 to receive empirical prophylaxis with oral ciprofloxacin (control group [CG]) or culture-based prophylaxis (intervention group [IG]). Primary outcome was any infectious complication within 7 days after biopsy. Secondary outcomes were infectious complications within 30 days, and bacteremia and bacteriuria within 7 and 30 days postbiopsy. For primary outcome analysis, the χ2 test stratified for hospitals was used. Trial registration number: NCT03228108.

RESULTS

Data from 1288 patients (83.7%) were available for analysis (CG, 652; IG, 636). Infection rates within 7 days postbiopsy were 4.3% (n = 28) (CG) and 2.5% (n = 16) (IG) (P value = .08; reduction: -1.8%; 95% confidence interval, -.004 to .040). Ciprofloxacin-resistant bacteria were detected in 15.2% (n = 1288). In the CG, the presence of ciprofloxacin-resistant rectal flora resulted in a 6.2-fold higher risk of early postbiopsy infection.

CONCLUSIONS

Our study supports the use of culture-based prophylaxis to reduce infectious complications after transrectal PB. Despite adequate prophylaxis, postbiopsy infections can still occur. Therefore, culture-based prophylaxis must be weighed against other strategies that could reduce postbiopsy infections. Clinical Trials Registration. NCT03228108.