External validation of the RSNA 2020 pulmonary embolism detection challenge winning deep learning algorithm.

PURPOSE: To evaluate the diagnostic performance and generalizability of the winning DL algorithm of the RSNA 2020 PE detection challenge to a local population using CTPA data from two hospitals. MATERIALS AND METHODS: Consecutive CTPA images from patients referred for suspected PE were retrospectively analysed. The winning RSNA 2020 DL algorithm was retrained on the RSNA-STR Pulmonary Embolism CT (RSPECT) dataset. The algorithm was tested in hospital A on multidetector CT (MDCT) images of 238 patients and in hospital B on spectral detector CT (SDCT) and virtual monochromatic images (VMI) of 114 patients. The output of the DL algorithm was compared with a reference standard, which included a consensus reading by at least two experienced cardiothoracic radiologists for both hospitals. Areas under the receiver operating characteristic curve (AUCs) were calculated. Sensitivity and specificity were determined using the maximum Youden index. RESULTS: According to the reference standard, PE was present in 73 patients (30.7%) in hospital A and 33 patients (29.0%) in hospital B. For the DL algorithm the AUC was 0.96 (95% CI 0.92-0.98) in hospital A, 0.89 (95% CI 0.81-0.94) for conventional reconstruction in hospital B and 0.87 (95% CI 0.80-0.93) for VMI. CONCLUSION: The RSNA 2020 pulmonary embolism detection on CTPA challenge winning DL algorithm, retrained on the RSPECT dataset, showed high diagnostic accuracy on MDCT images. A somewhat lower performance was observed on SDCT images, which suggest additional training on novel CT technology may improve generalizability of this DL algorithm.

An artificial intelligence algorithm for pulmonary embolism detection on polychromatic computed tomography: performance on virtual monochromatic images.

OBJECTIVES: Virtual monochromatic images (VMI) are increasingly used in clinical practice as they improve contrast-to-noise ratio. However, due to their different appearances, the performance of artificial intelligence (AI) trained on conventional CT images may worsen. The goal of this study was to assess the performance of an established AI algorithm trained on conventional polychromatic computed tomography (CT) images (CPI) to detect pulmonary embolism (PE) on VMI. METHODS: Paired 60 kiloelectron volt (keV) VMI and CPI of 114 consecutive patients suspected of PE, obtained with a detector-based spectral CT scanner, were retrospectively analyzed by an established AI algorithm. The CT pulmonary angiography (CTPA) were classified as positive or negative for PE on a per-patient level. The reference standard was established using a comprehensive method that combined the evaluation of the attending radiologist and three experienced cardiothoracic radiologists aided by two different detection tools. Sensitivity, specificity, positive and negative predictive values and likelihood ratios of the algorithm on VMI and CPI were compared. RESULTS: The prevalence of PE according to the reference standard was 35.1% (40 patients). None of the diagnostic accuracy measures of the algorithm showed a significant difference between CPI and VMI. Sensitivity was 77.5% (95% confidence interval (CI) 64.6-90.4%) and 85.0% (73.9-96.1%) (p = 0.08) on CPI and VMI respectively and specificity 96.0% (91.4-100.0%) and 94.6% (89.4-99.7%) (p = 0.32). CONCLUSIONS: Diagnostic performance of the AI algorithm that was trained on CPI did not drop on VMI, which is reassuring for its use in clinical practice. CLINICAL RELEVANCE STATEMENT: A commercially available AI algorithm, trained on conventional polychromatic CTPA, could be safely used on virtual monochromatic images. This supports the sustainability of AI-aided detection of PE on CT despite ongoing technological advances in medical imaging, although monitoring in daily practice will remain important. KEY POINTS: • Diagnostic accuracy of an AI algorithm trained on conventional polychromatic images to detect PE did not drop on virtual monochromatic images. • Our results are reassuring as innovations in hardware and reconstruction in CT are continuing, whilst commercial AI algorithms that are trained on older generation data enter healthcare.

Diffusion-weighted MRI with deep learning for visualizing treatment results of MR-guided HIFU ablation of uterine fibroids.

OBJECTIVES: No method is available to determine the non-perfused volume (NPV) repeatedly during magnetic resonance-guided high-intensity focused ultrasound (MR-HIFU) ablations of uterine fibroids, as repeated acquisition of contrast-enhanced T1-weighted (CE-T1w) scans is inhibited by safety concerns. The objective of this study was to develop and test a deep learning-based method for translation of diffusion-weighted imaging (DWI) into synthetic CE-T1w scans, for monitoring MR-HIFU treatment progression. METHODS: The algorithm was retrospectively trained and validated on data from 33 and 20 patients respectively who underwent an MR-HIFU treatment of uterine fibroids between June 2017 and January 2019. Postablation synthetic CE-T1w images were generated by a deep learning network trained on paired DWI and reference CE-T1w scans acquired during the treatment procedure. Quantitative analysis included calculation of the Dice coefficient of NPVs delineated on synthetic and reference CE-T1w scans. Four MR-HIFU radiologists assessed the outcome of MR-HIFU treatments and NPV ratio based on the synthetic and reference CE-T1w scans. RESULTS: Dice coefficient of NPVs was 71% (± 22%). The mean difference in NPV ratio was 1.4% (± 22%) and not statistically significant (p = 0.79). Absolute agreement of the radiologists on technical treatment success on synthetic and reference CE-T1w scans was 83%. NPV ratio estimations on synthetic and reference CE-T1w scans were not significantly different (p = 0.27). CONCLUSIONS: Deep learning-based synthetic CE-T1w scans derived from intraprocedural DWI allow gadolinium-free visualization of the predicted NPV, and can potentially be used for repeated gadolinium-free monitoring of treatment progression during MR-HIFU therapy for uterine fibroids. KEY POINTS: • Synthetic CE-T1w scans can be derived from diffusion-weighted imaging using deep learning. • Synthetic CE-T1w scans may be used for visualization of the NPV without using a contrast agent directly after MR-HIFU ablations of uterine fibroids.

Magnetic Resonance Imaging-targeted Prostate Biopsy Compared with Systematic Prostate Biopsy in Biopsy-naïve Patients with Suspected Prostate Cancer.

Background: It remains uncertain whether transrectal ultrasound (TRUS)-guided systematic biopsies can be omitted and rely solely on multiparametric magnetic resonance imaging-targeted biopsies (MRI-TBx) in biopsy-naïve men suspected of prostate cancer (PCa). Objective: To compare PCa detection in biopsy-naïve men between systematic biopsy and MRI-TBx. Design setting and participants: A prospective cohort study was conducted in a Dutch teaching hospital. Consecutive patients with suspected PCa, no history of biopsy, and no clinical suspicion of metastasis underwent both TRUS-guided systematic biopsies and MRI-TBx by multiparametric magnetic resonance imaging (mpMRI)-ultrasound fusion, including sham biopsies in case of negative mpMRI. Outcome measurements and statistical analysis: Clinically significant PCa (csPCa), defined as group ≥2 on the International Society of Urological Pathology grading, was detected. Results and limitations: The overall prevalence of csPCa, irrespective of biopsy technique, was 37.4% (132/353) in our population. MRI-TBx were performed in 263/353 (74.5%) patients with suspicious mpMRI (Prostate Imaging Reporting and Data System [PI-RADS] ≥3). The detection rates for csPCa were 39.5% for MRI-TBx and 42.9% for systematic biopsies. The added values, defined as the additional percentages of patients with csPCa detected by adding one biopsy technique, were 8.7% for the systematic biopsies and 5.3% for MRI-TBx. In patients with nonsuspicious mpMRI, five cases (6%) of csPCa were found by systematic biopsies. Conclusions: This study in biopsy-naïve patients suspected for PCa showed that systematic biopsies have added value to MRI-TBx alone in patients with mpMRI PI-RADS >2. Patient summary: We studied magnetic resonance imaging (MRI)-guided prostate biopsy for diagnosing prostate cancer and compared it with the standard method of prostate biopsy. Standard systematic biopsies cannot be omitted in patients with suspicious MRI, as they add to the detection of significant prostate cancer.

Increased MR-guided high intensity focused ultrasound (MR-HIFU) sonication efficiency of uterine fibroids after carbetocin administration.

Purpose: We investigated whether administration of the long-acting uterus stimulant carbetocin increased intra-subject sonication efficiency during Magnetic Resonance image guided High Intensity Focused Ultrasound (MR-HIFU) treatment of uterine fibroids. Method: In this prospective cohort study, thirty women with symptomatic uterine fibroids undergoing MR-HIFU treatment were included between January 2018 and January 2019. Treatment started with three sonications on one side of the uterine fibroid. Subsequently, one ampoule of 1 mL carbetocin (100 µg/mL) was administered intravenously and treatment continued with three sonications on the other side of the uterine fibroid. We compared the intra-subject sonication efficiency, in terms of Energy Efficiency Factor (EEF), thermal dose volume and sonication time to ablate one cm3 of fibroid tissue, before and after carbetocin administration. Adverse events that occurred within 30 min after carbetocin administration were recorded. Results: Sonication efficiency improved after carbetocin administration as indicated by a significant decrease in EEF and sonication time (p = 0.006 and p = 0.001 respectively), and a significant increase in thermal dose volume reached (p = <0.001). Five women (16.7%) experienced temporary tachycardia, one women in combination with headache, within 30 min after carbetocin administration. Conclusion: Administration of the long-acting uterus stimulant carbetocin improved the MR-HIFU treatment intra-subject sonication efficiency in women with symptomatic uterine fibroids.

Combining radiotherapy and focused ultrasound for pain palliation of cancer induced bone pain; a stage I/IIa study according to the IDEAL framework.

BACKGROUND: Cancer induced bone pain (CIBP) strongly interferes with patient's quality of life. Currently, the standard of care includes external beam radiotherapy (EBRT), resulting in pain relief in approximately 60% of patients. Magnetic Resonance guided High Intensity Focused Ultrasound (MR-HIFU) is a promising treatment modality for CIBP. METHODS: A single arm, R-IDEAL stage I/IIa study was conducted. Patients presenting at the department of radiation oncology with symptomatic bone metastases in the appendicular skeleton, as well as in the sacrum and sternum were eligible for inclusion. All participants underwent EBRT, followed by MR-HIFU within 4 days. Safety and feasibility were assessed, and pain scores were monitored for 4 weeks after completing the combined treatment. RESULTS: Six patients were enrolled. Median age was 67 years, median lesion diameter was 56,5 mm. In all patients it was logistically possible to plan and perform the MR-HIFU treatment within 4 days after EBRT. All patients tolerated the combined procedure well. Pain response was reported by 5 out of 6 patients at 7 days after completion of the combined treatment, and stabilized on 60% at 4 weeks follow up. No treatment related serious adverse events occurred. CONCLUSION: This is the first study to combine EBRT with MR-HIFU. Our results show that combined EBRT and MR-HIFU in first-line treatment of CIBP is safe and feasible, and is well tolerated by patients. Superiority over standard EBRT, in terms of (time to) pain relief and quality of life need to be evaluated in comparative (randomized) study.

Development and clinical evaluation of a 3-step modified manipulation protocol for MRI-guided high-intensity focused ultrasound of uterine fibroids.

OBJECTIVES: The clinical applicability of magnetic resonance image-guided high-intensity focused ultrasound (MR-HIFU) treatment of uterine fibroids is often limited due to inaccessible fibroids or bowel interference. The aim of this study was to implement a newly developed 3-step modified manipulation protocol and to evaluate its influence on the number of eligible women and treatment failure rate. METHODS: From June 2016 to June 2018, 165 women underwent a screening MRI examination, 67 women of whom were consecutively treated with MR-HIFU at our institution. Group 1 (n = 20) was treated with the BRB manipulation protocol which consisted of sequential applications of urinary bladder filling, rectal filling, and urinary bladder emptying. Group 2 (n = 47) was treated using the 3-step modified manipulation protocol which included (1) the BRB maneuver with adjusted rectal filling by adding psyllium fibers to the solution; (2) Trendelenburg position combined with bowel massage; (3) the manual uterine manipulation (MUM) method for uterine repositioning. A comparison was made between the two manipulation protocols to evaluate differences in safety, the eligibility percentage, and treatment failure rate due to unsuccessful manipulation. RESULTS: After implementing the 3-step modified manipulation protocol, our ineligibility rate due to bowel interference or inaccessible fibroids decreased from 18% (16/88) to 0% (0/77). Our treatment failure rate due to unsuccessful manipulation decreased from 20% (4/20) to 2% (1/47). There were no thermal complications to the bowel or uterus. CONCLUSIONS: Implementation of the 3-step modified manipulation protocol during MR-HIFU therapy of uterine fibroids improved the eligibility percentage and reduced the treatment failure rate. TRIAL REGISTRATION: Registry number NL56182.075.16 KEY POINTS: • A newly developed 3-step modified manipulation protocol was successfully implemented without the occurrence of thermal complication to the bowel or uterus. • The 3-step modified manipulation protocol increased our eligibility percentage for MR-HIFU treatment of uterine fibroids. • The 3-step modified manipulation protocol reduced our treatment failure rate for MR-HIFU treatment of uterine fibroids.

Emphysema quantification using chest CT: influence of radiation dose reduction and reconstruction technique.

BACKGROUND: Computed tomography (CT) emphysema quantification is affected by both radiation dose (i.e. image noise) and reconstruction technique. At reduced dose, filtered back projection (FBP) results in an overestimation of the amount of emphysema due to higher noise levels, while the use of iterative reconstruction (IR) can result in an underestimation due to reduced noise. The objective of this study was to determine the influence of dose reduction and hybrid IR (HIR) or model-based IR (MIR) on CT emphysema quantification. METHODS: Twenty-two patients underwent inspiratory chest CT scan at routine radiation dose and at 45%, 60% and 75% reduced radiation dose. Acquisitions were reconstructed with FBP, HIR and MIR. Emphysema was quantified using the 15th percentile of the attenuation curve and the percentage of voxels below -950 HU. To determine whether the use of a different percentile or HU threshold is more accurate at reduced dose levels and with IR, additional measurements were performed using different percentiles and HU thresholds to determine the optimal combination. RESULTS: Dose reduction resulted in a significant overestimation of emphysema, while HIR and MIR resulted in an underestimation. Lower HU thresholds with FBP at reduced dose and higher HU thresholds with HIR and MIR resulted in emphysema percentages comparable to the reference. The 15th percentile quantification method showed similar results as the HU threshold method. CONCLUSIONS: This within-patients study showed that CT emphysema quantification is significantly affected by dose reduction and IR. This can potentially be solved by adapting commonly used thresholds.

Mandatory adherence to diagnostic protocol increases the yield of CTPA for pulmonary embolism.

OBJECTIVES: To determine if mandatory adherence to a diagnostic protocol increases the rate of computed tomography pulmonary angiographies (CTPAs) positive for pulmonary embolism (PE)-the so-called diagnostic yield. Further, we aim to identify factors associated with this diagnostic yield. METHODS: We included all patients with suspected PE requiring CTPA from 9 January 2014 t0 3 June 2014. The requesting physicians were forced to follow diagnostic workup for PE by calculating a Wells score and, if necessary, determining D-dimer level. The percentage of positive CTPA scans was calculated and compared with our previous cohort (Walen et al. Insights Imaging 2014;5(2):231-236). Odds ratios were calculated as a measure of association between dichotomous variables and CTPA findings. RESULTS: Of 250 scans, 74 were positive (29.6 % [95 % CI, 24.3-35.5 %]) and 175 were negative (70 %). The percentage positive scans increased with 6.6 % and the percentage negative scans decreased with 3.1 %. This change was statistically significant (p = 0.001). Independent clinical predictors of diagnostic yield were previous deep venous thrombosis (DVT) (OR, 3.22; p = 0.013) and clinical signs of DVT (OR, 2.71; p = 0.012). Chronic obstructive pulmonary disease (COPD) was negatively associated with PE (OR, 0.33; p = 0.045). CONCLUSIONS: This study shows that mandatory adherence to a diagnostic protocol increases the yield of CTPA for PE in our centre. MAIN MESSAGES: • Mandatory adherence to diagnostic protocol increases the yield of CTPA for PE • Previous DVT and signs of DVT were associated with a higher yield • No patients with a low Wells score and a low D-dimer had PE.

[A patient with bilateral microcalcifications on mammography]. / Bilaterale microcalcificaties op het mammogram.

A 25-year-old woman with systemic lupus erythematosus presented with multiple palpable masses in both breasts. On mammography multiple, coarse, heterogeneous, microcalcificatons were seen. Histopathologic examination showed inflammation, fibrosis, fat necrosis and dystrophic microcalcifications. Based on these findings the patient was diagnosed with lupus mastitis.

Breast MRI in nonpalpable breast lesions: a randomized trial with diagnostic and therapeutic outcome - MONET - study.

BACKGROUND: In recent years there has been an increasing interest in MRI as a non-invasive diagnostic modality for the work-up of suspicious breast lesions. The additional value of Breast MRI lies mainly in its capacity to detect multicentric and multifocal disease, to detect invasive components in ductal carcinoma in situ lesions and to depict the tumor in a 3-dimensional image. Breast MRI therefore has the potential to improve the diagnosis and provide better preoperative staging and possibly surgical care in patients with breast cancer. The aim of our study is to assess whether performing contrast enhanced Breast MRI can reduce the number of surgical procedures due to better preoperative staging and whether a subgroup of women with suspicious nonpalpable breast lesions can be identified in which the combination of mammography, ultrasound and state-of-the-art contrast-enhanced Breast MRI can provide a definite diagnosis. METHODS/DESIGN: The MONET - study (MR mammography Of Nonpalpable BrEast Tumors) is a randomized controlled trial with diagnostic and therapeutic endpoints. We aim to include 500 patients with nonpalpable suspicious breast lesions who are referred for biopsy. With this number of patients, the expected 12% reduction in surgical procedures due to more accurate preoperative staging with Breast MRI can be detected with a high power (90%). The secondary outcome is the positive and negative predictive value of contrast enhanced Breast MRI. If the predictive values are deemed sufficiently close to those for large core biopsy then the latter, invasive, procedure could possibly be avoided in some women. The rationale, study design and the baseline characteristics of the first 100 included patients are described. TRIAL REGISTRATION: Study protocol number NCT00302120.