Reduction of ADC bias in diffusion MRI with deep learning-based acceleration: A phantom validation study at 3.0 T.

PURPOSE: Further acceleration of DWI in diagnostic radiology is desired but challenging mainly due to low SNR in high b-value images and associated bias in quantitative ADC values. Deep learning-based reconstruction and denoising may provide a solution to address this challenge. METHODS: The effects of SNR reduction on ADC bias and variability were investigated using a commercial diffusion phantom and numerical simulations. In the phantom, performance of different reconstruction methods, including conventional parallel (SENSE) imaging, compressed sensing (C-SENSE), and compressed SENSE acceleration with an artificial intelligence deep learning-based technique (C-SENSE AI), was compared at different acceleration factors and flip angles using ROI-based analysis. ADC bias was assessed by Lin's Concordance correlation coefficient (CCC) followed by bootstrapping to calculate confidence intervals (CI). ADC random measurement error (RME) was assessed by the mean coefficient of variation (CV¯) and non-parametric statistical tests. RESULTS: The simulations predicted increasingly negative bias and loss of precision towards lower SNR. These effects were confirmed in phantom measurements of increasing acceleration, for which CCC decreased from 0.947 to 0.279 and CV¯ increased from 0.043 to 0.439, and of decreasing flip angle, for which CCC decreased from 0.990 to 0.063 and CV¯ increased from 0.037 to 0.508. At high acceleration and low flip angle, C-SENSE AI reconstruction yielded best denoised ADC maps. For the lowest investigated flip angle, CCC = {0.630, 0.771 and 0.987} and CV¯={0.508, 0.426 and 0.254} were obtained for {SENSE, C-SENSE, C-SENSE AI}, the improvement by C-SENSE AI being significant as compared to the other methods (CV: p = 0.033 for C-SENSE AI vs. C-SENSE and p < 0.001 for C-SENSE AI vs. SENSE; CCC: non-overlapping CI between reconstruction methods). For the highest investigated acceleration factor, CCC = {0.479,0.926,0.960} and CV¯={0.519,0.119,0.118} were found, confirming the reduction of bias and RME by C-SENSE AI as compared to C-SENSE (by trend) and to SENSE (CV: p < 0.001; CCC: non-overlapping CI). CONCLUSION: ADC bias and random measurement error in DWI at low SNR, typically associated with scan acceleration, can be effectively reduced by deep-learning based C-SENSE AI reconstruction.

Targeted multiparametric magnetic resonance imaging/transrectal ultrasound-guided (mpMRI/TRUS) fusion prostate biopsy versus systematic random prostate biopsy: A comparative real-life study.

BACKGROUND: Patients with suspected prostate cancer usually undergo transrectal ultrasound-guided (TRUS) systematic biopsy, which can miss relevant prostate cancers and lead to overtreatment. AIMS: The aim of this study was to evaluate the detection rate for prostate cancer in MR-guided targeted biopsy (TB) and systematic biopsy (SB) in comparison with mpMRI of the prostate. METHODS AND RESULTS: Three hundred and eight men who underwent mpMRI due to elevated PSA values between 2015 and 2020 were studied at university hospital Aachen, Germany. MRI-images were divided into cohorts with suspicious findings (PI-RADS ≥ 3) and negative findings (PI-RADS < 3). In patients with PI-RADS ≥ 3 TB combined with SB was performed. A part of this group underwent RP subsequently. In patients with PI-RADS < 3 and clinical suspicion SB was performed. In the PI-RADS ≥ 3 group (n = 197), TB combined with SB was performed in 194 cases. Three cases were lost to follow-up. Biopsy yielded 143 positive biopsies and 51 cases without carcinoma. TB detected 71% (102/143) and SB 98% (140/143) of the overall 143 carcinoma. Overall, 102 carcinomas were detected by TB, hereof 66% (67/102) clinically significant (Gleason ≥ 3+4) and 34% (35/102) clinically insignificant carcinoma (Gleason 3+3). SB detected 140 carcinomas, hereof 64% (90/140) csPCA and 36% (50/140) nsPCA. Forty-one of the overall 143 detected carcinoma were only found by SB, hereof 46% (19/41) csPCA and 54% (22/41) nsPCA. Tumor locations overlapped in 44% (63/143) between TB and SB. In 25% (36/143), SB detected additional tumor foci outside the target lesions. 70/143 patients subsequently underwent RP. The detection of tumor foci was congruent between mpMRI and prostatectomy specimen in 79% (55/70) of cases. Tumor foci were mpMRI occult in 21% (15/70) of cases. In the group with negative mpMRI (n = 111), biopsy was performed in 81 cases. Gleason ≥ 3+4 carcinoma was detected in 7% and Gleason 3+3 in 24% cases. CONCLUSION: There was a notable number of cases in which SB detected tumor foci that were mpMRI occult and could have been missed by TB alone. Therefore, additional systematic random biopsy is still required. A supplemental random biopsy should be considered depending on the overall clinical suspicion in negative mpMRI.

Encrypted federated learning for secure decentralized collaboration in cancer image analysis.

Artificial intelligence (AI) has a multitude of applications in cancer research and oncology. However, the training of AI systems is impeded by the limited availability of large datasets due to data protection requirements and other regulatory obstacles. Federated and swarm learning represent possible solutions to this problem by collaboratively training AI models while avoiding data transfer. However, in these decentralized methods, weight updates are still transferred to the aggregation server for merging the models. This leaves the possibility for a breach of data privacy, for example by model inversion or membership inference attacks by untrusted servers. Somewhat-homomorphically-encrypted federated learning (SHEFL) is a solution to this problem because only encrypted weights are transferred, and model updates are performed in the encrypted space. Here, we demonstrate the first successful implementation of SHEFL in a range of clinically relevant tasks in cancer image analysis on multicentric datasets in radiology and histopathology. We show that SHEFL enables the training of AI models which outperform locally trained models and perform on par with models which are centrally trained. In the future, SHEFL can enable multiple institutions to co-train AI models without forsaking data governance and without ever transmitting any decryptable data to untrusted servers.

Denoising diffusion probabilistic models for 3D medical image generation.

Recent advances in computer vision have shown promising results in image generation. Diffusion probabilistic models have generated realistic images from textual input, as demonstrated by DALL-E 2, Imagen, and Stable Diffusion. However, their use in medicine, where imaging data typically comprises three-dimensional volumes, has not been systematically evaluated. Synthetic images may play a crucial role in privacy-preserving artificial intelligence and can also be used to augment small datasets. We show that diffusion probabilistic models can synthesize high-quality medical data for magnetic resonance imaging (MRI) and computed tomography (CT). For quantitative evaluation, two radiologists rated the quality of the synthesized images regarding "realistic image appearance", "anatomical correctness", and "consistency between slices". Furthermore, we demonstrate that synthetic images can be used in self-supervised pre-training and improve the performance of breast segmentation models when data is scarce (Dice scores, 0.91 [without synthetic data], 0.95 [with synthetic data]).

Artificial Intelligence for Clinical Interpretation of Bedside Chest Radiographs.

Background Supine chest radiography for bedridden patients in intensive care units (ICUs) is one of the most frequently ordered imaging studies worldwide. Purpose To evaluate the diagnostic performance of a neural network-based model that is trained on structured semiquantitative radiologic reports of bedside chest radiographs. Materials and Methods For this retrospective single-center study, children and adults in the ICU of a university hospital who had been imaged using bedside chest radiography from January 2009 to December 2020 were reported by using a structured and itemized template. Ninety-eight radiologists rated the radiographs semiquantitatively for the severity of disease patterns. These data were used to train a neural network to identify cardiomegaly, pulmonary congestion, pleural effusion, pulmonary opacities, and atelectasis. A held-out internal test set (100 radiographs from 100 patients) that was assessed independently by an expert panel of six radiologists provided the ground truth. Individual assessments by each of these six radiologists, by two nonradiologist physicians in the ICU, and by the neural network were compared with the ground truth. Separately, the nonradiologist physicians assessed the images without and with preliminary readings provided by the neural network. The weighted Cohen κ coefficient was used to measure agreement between the readers and the ground truth. Results A total of 193 566 radiographs in 45 016 patients (mean age, 66 years ± 16 [SD]; 61% men) were included and divided into training (n = 122 294; 64%), validation (n = 31 243; 16%), and test (n = 40 029; 20%) sets. The neural network exhibited higher agreement with a majority vote of the expert panel (κ = 0.86) than each individual radiologist compared with the majority vote of the expert panel (κ = 0.81 to ≤0.84). When the neural network provided preliminary readings, the reports of the nonradiologist physicians improved considerably (aided vs unaided, κ = 0.87 vs 0.79, respectively; P < .001). Conclusion A neural network trained with structured semiquantitative bedside chest radiography reports allowed nonradiologist physicians improved interpretations compared with the consensus reading of expert radiologists. © RSNA, 2022 Supplemental material is available for this article. See also the editorial by Wielpütz in this issue.

Bone Mineral Density Is a Predictor of Mortality in Female Patients with Cholangiocellular Carcinoma Undergoing Palliative Treatment.

BACKGROUND: Cholangiocellular adenocarcinoma (CCA) is a rare and aggressive malignancy originating from the bile ducts. Its general prognosis is poor as therapeutic options are limited. Many patients present with advanced stages of disease, and palliative chemotherapy remains the only treatment option. Prognostic markers to assess the outcome of chemotherapeutic treatment in CCA are limited. We therefore evaluated bone mineral density (BMD) as a prognostic tool in patients with advanced CCA. PATIENTS AND METHODS: We included 75 patients with advanced CCA that were treated at our academic tumor center. Prior to treatment, bone mineral density was analyzed at the first lumbar vertebra using routine CT scans in the venous phase and the local PACS (IntelliSpace PACS, Philips, Amsterdam, The Netherlands). RESULTS: BMD was not significantly different between male and female patients but decreased with age. Patients with BMD above 167 HU have a significantly improved overall survival (474 days vs. 254 days; log-rank X2(1) = 6.090; p = 0.014). The prognostic value of BMD was confirmed using univariate (HR 2.313 (95%CI: 1.170-4.575); p = 0.016) and multivariate (HR 4.143 (95%CI: 1.197-14.343); p = 0.025) Cox regression analyses. Subgroup analysis revealed that the prognostic value of BMD was only present in female patients and not in male patients, suggesting sex-specific differences. CONCLUSIONS: Our data suggest that BMD is a valuable, easily accessible, and independent prognostic marker for overall survival in patients with advanced CCA. Furthermore, subgroup analysis showed the sex specificity of this marker, which demonstrated relevance only in female patients.

Possible Alterations of Imaging Patterns in Computed Tomography for Delta-VOC of SARS-CoV-2.

BACKGROUND: So far, typical findings for COVID-19 in computed tomography (CT) have been described as bilateral, multifocal ground glass opacities (GGOs) and consolidations, as well as intralobular and interlobular septal thickening. On the contrary, round consolidations with the halo sign are considered uncommon and are typically found in fungal infections, such as invasive pulmonary aspergillosis. The authors recently observed several patients with COVID-19 pneumonia presenting with round, multifocal consolidations accompanied by a halo sign. As this may indicate alterations of CT morphology based on the virus variant, the aim of this study was to investigate this matter in more detail. METHODS: 161 CT scans of patients with confirmed SARS-CoV-2 infection (RT-PCR within 2 days of CT) examined between January 2021 and September 15, 2021 were included. Follow-up examinations, patients with invasive ventilation at the time of CT, and patients with insufficient virus typing for variants of concern (VOC) were excluded. CT scans were assessed for vertical and axial distribution of pulmonary patterns, degree of involvement, uni- vs. bilaterality, reticulations, and other common findings. The mean density of representative lesions was assessed in Hounsfield units. Results were compared using Mann-Whitney U-tests, Student's t-rests, descriptive statistics, and Fisher's exact tests. RESULTS: 75 patients did not meet the inclusion criteria. Therefore, 86/161 CT scans of unique patients were analyzed. PCR VOC testing confirmed manifestation of the Delta-VOC SARS-CoV-2 in 22 patients, 39 patients with Alpha-VOC and the remaining 25 patients with Non-VOC SARS-CoV-2 infections. Three patients with the Delta-VOC demonstrated multiple pulmonary masses or nodules with surrounding halo sign, whereas no patients with either Alpha-VOC (p = 0.043) or non-VOC (p = 0.095) demonstrated these findings. All three patients were admitted to normal wards and had no suspicion of a pulmonary co-infection. Patients with Delta-VOC were less likely to have ground glass opacities compared to Alpha-VOC (7/22 or 31.8 % vs. 4/39 or 10.3 %; p < 0.001), whereas a significant difference has not been observed between Delta-VOC and non-VOC (5/25 or 20 %; p = 0.348). The mean representative density of lesions did not show significant differences between the studied cohorts. CONCLUSION: In this study 3 out of 22 patients (13.6 %) with Delta-VOC presented with bilateral round pulmonary masses or nodules with surrounding halo signs, which has not been established as a notable imaging pattern in COVID-19 pneumonia yet. Compared to the other cohorts, a lesser percentage of patients with Delta-VOC presented with ground glass opacities. Based on these results Delta-VOC might cause a divergence in CT-morphologic phenotype. KEY POINTS: · Until recently, CT-morphologic signs of COVID-19 pneumonia have been presumed to be uncontroversially understood. Yet, recently the authors observed diverging pulmonary alterations in patients infected with Delta-VOC.. · These imaging alterations included round pulmonary masses or nodules with surrounding halo sign.. · These imaging alterations have not yet been established as typical for COVID-19 pneumonia, yet.. · Based on these results, Delta-VOC could impose a divergence of CT-morphologic phenotype.. CITATION FORMAT: · Yüksel C, Sähn M, Kleines M et al. Possible Alterations of Imaging Patterns in Computed Tomography for Delta-VOC of SARS-CoV-2 . Fortschr Röntgenstr 2022; 194: 1229 - 1241.

Reliability as a Precondition for Trust-Segmentation Reliability Analysis of Radiomic Features Improves Survival Prediction.

Machine learning results based on radiomic analysis are often not transferrable. A potential reason for this is the variability of radiomic features due to varying human made segmentations. Therefore, the aim of this study was to provide comprehensive inter-reader reliability analysis of radiomic features in five clinical image datasets and to assess the association of inter-reader reliability and survival prediction. In this study, we analyzed 4598 tumor segmentations in both computed tomography and magnetic resonance imaging data. We used a neural network to generate 100 additional segmentation outlines for each tumor and performed a reliability analysis of radiomic features. To prove clinical utility, we predicted patient survival based on all features and on the most reliable features. Survival prediction models for both computed tomography and magnetic resonance imaging datasets demonstrated less statistical spread and superior survival prediction when based on the most reliable features. Mean concordance indices were Cmean = 0.58 [most reliable] vs. Cmean = 0.56 [all] (p < 0.001, CT) and Cmean = 0.58 vs. Cmean = 0.57 (p = 0.23, MRI). Thus, preceding reliability analyses and selection of the most reliable radiomic features improves the underlying model's ability to predict patient survival across clinical imaging modalities and tumor entities.

Automated major psoas muscle volumetry in computed tomography using machine learning algorithms.

PURPOSE: The psoas major muscle (PMM) volume serves as an opportunistic imaging marker in cross-sectional imaging datasets for various clinical applications. Since manual segmentation is time consuming, two different automated segmentation methods, a generative adversarial network architecture (GAN) and a multi-atlas segmentation (MAS), as well as a combined approach of both, were investigated in terms of accuracy of automated volumetrics in given CT datasets. MATERIALS AND METHODS: The bilateral PMM was manually segmented by a radiologist in 34 abdominal CT scans, resulting in 68 single 3D muscle segmentations as training data. Three different methods were tested for their ability to generate automated image segmentations: a GAN- and MAS-based approach and a combined approach of both methods (COM). Bilateral PMM volume (PMMV) was calculated in cm3 by each algorithm for every CT. Results were compared to the corresponding ground truth using the Dice similarity coefficient (DSC), Spearman's correlation coefficient and Wilcoxon signed-rank test. RESULTS: Mean PMMV was 239 ± 7.0 cm3 and 308 ± 9.6 cm3, 306 ± 9.5 cm3 and 243 ± 7.3 cm3 for the CNN, MAS and COM, respectively. Compared to the ground truth the CNN and MAS overestimated the PMMV significantly (+ 28.9% and + 28.0%, p < 0.001), while results of the COM were quite accurate (+ 0.7%, p = 0.33). Spearman's correlation coefficients were 0.38, 0.62 and 0.73, and the DSCs were 0.75 [95%CI: 0.56-0.88], 0.73 [95%CI: 0.54-0.85] and 0.82 [95%CI: 0.65-0.90] for the CNN, MAS and COM, respectively. CONCLUSION: The combined approach was able to efficiently exploit the advantages of both methods (GAN and MAS), resulting in a significantly higher accuracy in PMMV predictions compared to the isolated implementations of both methods. Even with the relatively small set of training data, the segmentation accuracy of this hybrid approach was relatively close to that of the radiologist.

Sarcopenia Predicts Cancer Mortality in Male but Not in Female Patients Undergoing Surgery for Cholangiocellular Carcinoma.

INTRODUCTION: Surgery represents the only curative treatment option for patients with cholangiocarcinoma. However, complete tumor resection requires extensive surgery in many patients, and it is still debated which patients represent the ideal candidates for such therapy in terms of overall survival. Sarcopenia has been associated with an adverse outcome for various malignancies, but its role in the context of patients undergoing tumor resection for cholangiocellular adenocarcinoma (CCA) is only poorly understood. Here, we evaluated the role of sarcopenia in the outcome of CCA patients undergoing radical tumor resection. METHODS: Pre-operative CT scans were used to assess the skeletal muscle index (L3SMI) as well as the psoas muscle index (L3PMI) in n = 76 patients receiving curative intended surgery for CCA. L3SMI and L3PMI were correlated with clinical and laboratory markers. RESULTS: Patients with a skeletal muscle index or psoas muscle index above an established ideal cut-off (54.26 and 1.685 cm2/m2) showed a significantly better overall survival in Kaplan-Meier Curve analyses (L3SMI: 1814 days (95% CI: 520-3108) vs. 467 days (95% CI: 225-709) days; log rank X2(1) = 7.18, p = 0.007; L3PMI: 608 days (95% CI: 297-919) vs. 87 days (95% CI: 33-141), log rank X2(1) = 18.71; p < 0.001). Notably, these findings, especially for L3PMI, were confirmed in univariate (L3SMI: HR 0.962 (0.936-0.989); p = 0.006; L3PMI: HR 0.529 (0.366-0.766); p ≤ 0.001) and multivariate Cox regression analyses. Further analyses revealed that the prognostic value of both L3SMI and L3PMI was restricted to male patients, while in female patients survival was independent of the individual muscle mass. CONCLUSION: Measurement of muscle mass from preoperative CT scans represents an easily obtainable tool to estimate patient prognosis following curative surgery. The prognostic value was restricted to male patients, while in female patients these parameters did not reflect the patient outcome.

3D-Printed Replica and Porcine Explants for Pre-Clinical Optimization of Endoscopic Tumor Treatment by Magnetic Targeting.

BACKGROUND: Animal models have limitations in cancer research, especially regarding anatomy-specific questions. An example is the exact endoscopic placement of magnetic field traps for the targeting of therapeutic nanoparticles. Three-dimensional-printed human replicas may be used to overcome these pitfalls. METHODS: We developed a transparent method to fabricate a patient-specific replica, allowing for a broad scope of application. As an example, we then additively manufactured the relevant organs of a patient with locally advanced pancreatic ductal adenocarcinoma. We performed experimental design investigations for a magnetic field trap and explored the best fixation methods on an explanted porcine stomach wall. RESULTS: We describe in detail the eight-step development of a 3D replica from CT data. To guide further users in their decisions, a morphologic box was created. Endoscopies were performed on the replica and the resulting magnetic field was investigated. The best fixation method to hold the magnetic field traps stably in place was the fixation of loops at the stomach wall with endoscopic single-use clips. CONCLUSIONS: Using only open access software, the developed method may be used for a variety of cancer-related research questions. A detailed description of the workflow allows one to produce a 3D replica for research or training purposes at low costs.

Sarcopenia and Myosteatosis as Prognostic Markers in Patients with Advanced Cholangiocarcinoma Undergoing Palliative Treatment.

BACKGROUND: Cholangiocarcinoma (CCA) represents the second most common primary liver cancer and is characterized by a very poor outcome, but reliable prognostic markers are largely missing. Sarcopenia, the progressive loss of muscle mass and strength, as well as myosteatosis have been associated with an unfavorable outcome in several clinical conditions, including cancer. Here, we evaluated the prognostic relevance of sarcopenia and myosteatosis using routine abdominal CT (computed tomography) scans in advanced stage CCA patients undergoing palliative treatment. METHODS: Routine abdominal CT scans were used to assess the skeletal muscle and the psoas muscle index (L3SMI/L3PMI) at the level of the third lumbar vertebra as radiological indices for sarcopenia as well as the mean skeletal muscle attenuation (MMA) as a surrogate for myosteatosis. Results were correlated with clinical data and outcomes. RESULTS: Using a calculated optimal cut-off value of 71.95 mm2/cm, CCA patients with an L3SMI value below this cut-off showed a significantly reduced median overall survival (OS) of only 250 days compared to 450 days in patients with a higher L3SMI. Moreover, the median OS of CCA patients with an L3PMI above 6345 mm2/cm was 552 days compared to 252 days in patients with a lower L3PMI. Finally, CCA patients with an MMA above 30.51 Hounsfield Units survived significantly longer (median OS: 430 days) compared to patients with an MMA value below this ideal cut-off (median OS: 215 days). The prognostic relevance of L3SMI, L3PMI, and MMA was confirmed in uni- and multivariate Cox regression analyses. CONCLUSION: Routine abdominal CT scans represent a unique opportunity to evaluate sarcopenia as well as myosteatosis in advanced CCA patients. We identified the L3SMI/L3PMI as well as the MMA as negative prognostic factors in CCA patients undergoing palliative therapy, arguing that the "opportunistic" evaluation of these parameters might yield important clinical information in daily routine.

A Radiomics Approach to Predict the Emergence of New Hepatocellular Carcinoma in Computed Tomography for High-Risk Patients with Liver Cirrhosis.

Liver cirrhosis poses a major risk for the development of hepatocellular carcinoma (HCC). This retrospective study investigated to what extent radiomic features allow the prediction of emerging HCC in patients with cirrhosis in contrast-enhanced computed tomography (CECT). A total of 51 patients with liver cirrhosis and newly detected HCC lesions (n = 82) during follow-up (FU-CT) after local tumor therapy were included. These lesions were not to have been detected by the radiologist in the chronologically prior CECT (PRE-CT). For training purposes, segmentations of 22 patients with liver cirrhosis but without HCC-recurrence were added. A total of 186 areas (82 HCCs and 104 cirrhotic liver areas without HCC) were analyzed. Using univariate analysis, four independent features were identified, and a multivariate logistic regression model was trained to classify the outlined regions as "HCC probable" or "HCC improbable". In total, 60/82 (73%) of segmentations with later detected HCC and 84/104 (81%) segmentations without HCC were classified correctly (AUC of 81%, 95% CI 74-87%), yielding a sensitivity of 72% (95% CI 57-83%) and a specificity of 86% (95% CI 76-96%). In conclusion, the model predicted the occurrence of new HCCs within segmented areas with an acceptable sensitivity and specificity in cirrhotic liver tissue in CECT.

Serum Perilipin 2 (PLIN2) Predicts Multiple Organ Dysfunction in Critically Ill Patients.

Perilipin 2 (PLIN2) is a lipid droplet protein with various metabolic functions. However, studies investigating PLIN2 in the context of inflammation, especially in systemic and acute inflammation, are lacking. Hence, we assessed the relevance of serum PLIN2 in critically ill patients. We measured serum PLIN2 serum in 259 critically ill patients (166 with sepsis) upon admission to a medical intensive care unit (ICU) compared to 12 healthy controls. A subset of 36 patients underwent computed tomography to quantify body composition. Compared to controls, serum PLIN2 concentrations were elevated in critically ill patients at ICU admission. Interestingly, PLIN2 independently indicated multiple organ dysfunction (MOD), defined as a SOFA score > 9 points, at ICU admission, and was also able to independently predict MOD after 48 h. Moreover, serum PLIN2 levels were associated with severe respiratory failure potentially reflecting a moribund state. However, PLIN2 was neither a predictor of ICU mortality nor did it reflect metabolic dysregulation. Conclusively, the first study assessing serum PLIN2 in critical illness proved that it may assist in risk stratification because it is capable of independently indicating MOD at admission and predicting MOD 48 h after PLIN2 measurement. Further evaluation regarding the underlying mechanisms is warranted.

Decreased Bone Mineral Density Is a Predictor of Poor Survival in Critically Ill Patients.

Alterations in bone mineral density (BMD) have been suggested as independent predictors of survival for several diseases. However, little is known about the role of BMD in the context of critical illness and intensive care medicine. We therefore evaluated the prognostic role of BMD in critically ill patients upon admission to an intensive care unit (ICU). Routine computed tomography (CT) scans of 153 patients were used to assess BMD in the first lumbar vertebra. Results were correlated with clinical data and outcomes. While median BMD was comparable between patients with and without sepsis, BMD was lower in patients with pre-existing arterial hypertension or chronic obstructive pulmonary disease. A low BMD upon ICU admission was significantly associated with impaired short-term ICU survival. Moreover, patients with baseline BMD < 122 HU had significantly impaired overall survival. The prognostic relevance of low BMD was confirmed in uni- and multivariate Cox-regression analyses including several clinicopathological parameters. In the present study, we describe a previously unrecognised association of individual BMD with short- and long-term outcomes in critically ill patients. Due to its easy accessibility in routine CT, BMD provides a novel prognostic tool to guide decision making in critically ill patients.

Advancing diagnostic performance and clinical usability of neural networks via adversarial training and dual batch normalization.

Unmasking the decision making process of machine learning models is essential for implementing diagnostic support systems in clinical practice. Here, we demonstrate that adversarially trained models can significantly enhance the usability of pathology detection as compared to their standard counterparts. We let six experienced radiologists rate the interpretability of saliency maps in datasets of X-rays, computed tomography, and magnetic resonance imaging scans. Significant improvements are found for our adversarial models, which are further improved by the application of dual-batch normalization. Contrary to previous research on adversarially trained models, we find that accuracy of such models is equal to standard models, when sufficiently large datasets and dual batch norm training are used. To ensure transferability, we additionally validate our results on an external test set of 22,433 X-rays. These findings elucidate that different paths for adversarial and real images are needed during training to achieve state of the art results with superior clinical interpretability.

CT-Based Prediction of Liver Function and Post-PVE Hypertrophy Using an Artificial Neural Network.

BACKGROUND: This study aimed to evaluate whether hypertrophy after portal vein embolization (PVE) and maximum liver function capacity (LiMAx) are predictable by an artificial neural network (ANN) model based on computed tomography (CT) texture features. METHODS: We report a retrospective analysis on 118 patients undergoing preoperative assessment by CT before and after PVE for subsequent extended liver resection due to a malignant tumor at RWTH Aachen University Hospital. The LiMAx test was carried out in a subgroup of 55 patients prior to PVE. Associations between CT texture features and hypertrophy as well as liver function were assessed by a multilayer perceptron ANN model. RESULTS: Liver volumetry showed a median hypertrophy degree of 33.9% (16.5-60.4%) after PVE. Non-response, defined as a hypertrophy grade lower than 25%, was found in 36.5% (43/118) of the cases. The ANN prediction of the hypertrophy response showed a sensitivity of 95.8%, specificity of 44.4% and overall prediction accuracy of 74.6% (p < 0.001). The observed median LiMAx was 327 (248-433) µg/kg/h and was strongly correlated with the predicted LiMAx (R2 = 0.89). CONCLUSION: Our study shows that an ANN model based on CT texture features is able to predict the maximum liver function capacity and may be useful to assess potential hypertrophy after performing PVE.

Can the predictive value of multiparametric MRI for prostate cancer be improved by a liquid biopsy with SelectMDx?

BACKGROUND: SelectMDx is a urinary biomarker test for determining prostate cancer risk. AIM: In a group of patients with a biopsy proven prostate cancer (PCa) who had undergone a multi parametric Magnetic Resonance Imaging (mpMRI) and urinary biomarker test with SelectMDx, we studied the additive value of SelectMDx to mpMRI and correlated that to the radical prostatectomy histology. METHODS AND RESULTS: Thirty-nine consecutive patients with a positive prostate biopsy were included in the study. They all had mpMRI and SelectMDx and underwent a radical prostatectomy. Overall, the mpMRI showed a PIRADS ≤3 lesion in seven cases out of the 39 patients. Significant lesions (PIRADS ≥4) were found in 32 cases (82%), that is, in 17 cases a PIRADS 5 lesion and in 15 cases a PIRADS 4 lesion. The mpMRI missed significant PCa in seven cases (18%) who had a PIRADS ≤3 lesion but had a significant PCa on final histology after RP. In our study, the positive predictive values of mpMRI were 97% and that of the SelectMDx was 100%. CONCLUSION: In this real-life selected group of consecutive patients with a confirmed positive PCa biopsy and available mpMRI, the liquid biopsy test with SelectMDx, did not provide an additional information about the PCa clinical significance. The addition of SelectMDx was only found valuable in those patients who had a very high-risk PCa (ie, GS ≥8) who had a positive SelectMDx test outcome despite of a negative mpMRI outcome.

CT-based determination of excessive visceral adipose tissue is associated with an impaired survival in critically ill patients.

OBJECTIVE: Obesity is a negative prognostic factor for various clinical conditions. In this observational cohort study, we evaluated a CT-based assessment of the adipose tissue distribution as a potential non-invasive prognostic parameter in critical illness. METHODS: Routine CT-scans upon admission to the intensive care unit (ICU) were used to analyze the visceral and subcutaneous adipose tissue areas at the 3rd lumbar vertebra in 155 patients. Results were correlated with various prognostic markers and both short-term- and overall survival. Multiple statistical tools were used for data analysis. RESULTS: We observed a significantly larger visceral adipose tissue area in septic patients compared to non-sepsis patients. Interestingly, patients requiring mechanical ventilation had a significantly higher amount of visceral adipose tissue correlating with the duration of mechanical ventilation. Moreover, both visceral and subcutaneous adipose tissue area significantly correlated with several laboratory markers. While neither the visceral nor the subcutaneous adipose tissue area was predictive for short-term ICU survival, patients with a visceral adipose tissue area above the optimal cut-off (241.4 cm2) had a significantly impaired overall survival compared to patients with a lower visceral adipose tissue area. CONCLUSIONS: Our study supports a prognostic role of the individual adipose tissue distribution in critically ill patients. However, additional investigations need to confirm our suggestion that routine CT-based assessment of adipose tissue distribution can be used to yield further information on the patients' clinical course. Moreover, future studies should address functional and metabolic analysis of different adipose tissue compartments in critical illness.