Diagnostic accuracy of texture analysis and machine learning for quantification of liver fibrosis in MRI: correlation with MR elastography and histopathology.

OBJECTIVES: To compare the diagnostic accuracy of texture analysis (TA)-derived parameters combined with machine learning (ML) of non-contrast-enhanced T1w and T2w fat-saturated (fs) images with MR elastography (MRE) for liver fibrosis quantification. METHODS: In this IRB-approved prospective study, liver MRIs of participants with suspected chronic liver disease who underwent liver biopsy between August 2015 and May 2018 were analyzed. Two readers blinded to clinical and histopathological findings performed TA. The participants were categorized into no or low-stage (0-2) and high-stage (3-4) fibrosis groups. Confusion matrices were calculated using a support vector machine combined with principal component analysis. The diagnostic accuracy of ML-based TA of liver fibrosis and MRE was assessed by area under the receiver operating characteristic curves (AUC). Histopathology served as reference standard. RESULTS: A total of 62 consecutive participants (40 men; mean age ± standard deviation, 48 ± 13 years) were included. The accuracy of TA and ML on T1w was 85.7% (95% confidence interval [CI] 63.7-97.0) and 61.9% (95% CI 38.4-81.9) on T2w fs for classification of liver fibrosis into low-stage and high-stage fibrosis. The AUC for TA on T1w was similar to MRE (0.82 [95% CI 0.59-0.95] vs. 0.92 [95% CI 0.71-0.99], p = 0.41), while the AUC for T2w fs was significantly lower compared to MRE (0.57 [95% CI 0.34-0.78] vs. 0.92 [95% CI 0.71-0.99], p = 0.008). CONCLUSION: Our results suggest that liver fibrosis can be quantified with TA-derived parameters of T1w when combined with a ML algorithm with similar accuracy compared to MRE. KEY POINTS: • Liver fibrosis can be categorized into low-stage fibrosis (0-2) and high-stage fibrosis (3-4) using texture analysis-derived parameters of T1-weighted images with a machine learning approach. • For the differentiation of low-stage fibrosis and high-stage fibrosis, the diagnostic accuracy of texture analysis on T1-weighted images combined with a machine learning algorithm is similar compared to MR elastography.

Assessment of hepatic fibrosis and inflammation with look-locker T1 mapping and magnetic resonance elastography with histopathology as reference standard.

PURPOSE: To compare the diagnostic performance of T1 mapping and MR elastography (MRE) for staging of hepatic fibrosis and grading inflammation with histopathology as standard of reference. METHODS: 68 patients with various liver diseases undergoing liver biopsy for suspected fibrosis or with an established diagnosis of cirrhosis prospectively underwent look-locker inversion recovery T1 mapping and MRE. T1 relaxation time and liver stiffness (LS) were measured by two readers. Hepatic fibrosis and inflammation were histopathologically staged according to a standardized fibrosis (F0-F4) and inflammation (A0-A2) score. For statistical analysis, independent t test, and Mann-Whitney U test and ROC analysis were performed, the latter to determine the performance of T1 mapping and MRE for fibrosis staging and inflammation grading, as compared to histopathology. RESULTS: Histopathological analysis diagnosed 9 patients with F0 (13.2%), 21 with F1 (30.9%), 11 with F2 (16.2%), 10 with F3 (14.7%), and 17 with F4 (25.0%). Both T1 mapping and MRE showed significantly higher values for patients with significant fibrosis (F0-1 vs. F2-4; T1 mapping p < 0.0001, MRE p < 0.0001) as well as for patients with severe fibrosis or cirrhosis (F0-2 vs. F3-4; T1 mapping p < 0.0001, MRE p < 0.0001). T1 values and MRE LS were significantly higher in patients with inflammation (A0 vs. A1-2, both p = 0.01). T1 mapping showed a tendency toward lower diagnostic performance without statistical significance for significant fibrosis (F2-4) (AUC 0.79 vs. 0.91, p = 0.06) and with a significant difference compared to MRE for severe fibrosis (F3-4) (AUC 0.79 vs. 0.94, p = 0.03). For both T1 mapping and MRE, diagnostic performance for diagnosing hepatic inflammation (A1-2) was low (AUC 0.72 vs. 0.71, respectively). CONCLUSION: T1 mapping is able to diagnose hepatic fibrosis, however, with a tendency toward lower diagnostic performance compared to MRE and thus may be used as an alternative to MRE for diagnosing hepatic fibrosis, whenever MRE is not available or likely to fail due to intrinsic factors of the patient. Both T1 mapping and MRE are probably not sufficient as standalone methods to diagnose hepatic inflammation with relatively low diagnostic accuracy.

Hypothermia induced alteration of repolarization - impact on acute and long-term outcome: a prospective cohort study.

BACKGROUND: The effects of target temperature management (TTM) on the heart aren't thoroughly studied yet. Several studies showed the prolongation of various ECG parameters including Tpeak-Tend-time under TTM. Our study's goal is to evaluate the acute and long-term outcome of these prolongations. METHODS: In this study we included patients with successful resuscitation after cardiac arrest who were admitted to the Charité Virchow Klinikum Berlin or the Heart and Vascular Centre of the Ruhr University Bochum between February 2006 and July 2013 (Berlin) or May 2014 to November 2015 (Bochum). For analysis, one ECG during TTM was recorded after reaching the target temperature (33-34 °C) or in the first 6 h of TTM. If possible, another ECG was taken after TTM. The patients were being followed until February 2016. Primary endpoint was ventricular arrhythmia during TTM, secondary endpoints were death and hospitalization due to cardiovascular diseases during follow-up. RESULTS: One hundred fifty-eight patients were successfully resuscitated in the study period of which 95 patients had usable data (e.g. ECGs without artifacts). During TTM significant changes for different parameters of ventricular de- and repolarization were noted: QRS (103.2 ± 23.7 vs. 95.3 ± 18.1; p = 0.003),QT (405.8 ± 76.4 vs. 373.8 ± 75.0; p = 0.01), QTc (474.9 ± 59.7 vs. 431.0 ± 56.8; p < 0.001), JT (302.8 ± 69.4 vs. 278.5 ± 75.2; p = 0.043), JTc (354.3 ± 60.2 vs. 318.7 ± 59.1; p = 0.001). 13.7% of the patients had ventricular arrhythmias during TTM, however these patients showed no difference regarding their ECG parameters in comparison to those were no ventricular arrhythmias occurred. We were able to follow 69 Patients over an average period of 35 ± 31 months. The 14 (21.5%) patients who died during the follow-up had significant prolongations of the TpTe-time in the ECGs without TTM (103.9 ± 47.2 vs. 75.8 ± 28.6; p = 0.023). CONCLUSION: Our results show a significant prolongation of ventricular repolarization during TH. However, there was no significant difference between the ECG parameters of those who developed a ventricular arrhythmia and those who did not. The temporary prolongation of the repolarization during TTM seems to be less important for the prognosis of the patient. Whereas the prolongation of the repolarization in the basal ECG is associated with a higher mortality in our study.

Where There's Smoke, There's Fire? Significance of Atrial Fibrillation in Young Patients.

BACKGROUND: Atrial fibrillation (AF), a disease of the elderly, occasionally occurs at younger age. Pathophysiology of AF in younger patients is diverse, including channelopathies and cardiomyopathies. Data on the significance and complications of AF in young patients are scarce. HYPOTHESIS: Atrial fibrillation is the first manifestation of cardiovascular disease (CVD) in young patients. METHODS: From 11 888 patients in a university hospital database, patients age ≤35 years were identified. A composite of stroke/transient ischemic attack, thromboembolic events, major bleeding, and death was the primary endpoint. Stroke/transient ischemic attack, thromboembolic events, major bleeding, death, AF during follow-up, diagnosis of arrhythmia other than AF, and new diagnosis of any CVD were secondary endpoints. Endpoints were compared between patients with and without comorbidities. RESULTS: We identified 124 patients (29.1± 5 years). Of those, 84 were followed over 48.4 ± 39.8 months. Comorbidities were present in 40.5%. Incidence of the primary endpoint was not different between the groups. Arrhythmias other than AF were more common in patients without comorbidities (36% vs 14.7%; P = 0.032). A supraventricular tachycardia (SVT) was found in 57.1% of patients who underwent electrophysiological testing and was treated with catheter ablation. None of those patients had AF during follow-up. CONCLUSIONS: Atrial fibrillation occurs in young patients with and without structural heart disease. Young AF patients without comorbidities rarely develop CVD during the first years after diagnosis. Yet, an SVT is found in a high proportion of young AF patients; AF may be first manifestation of SVT. Therefore, young patients should undergo further evaluation for SVT.

Performance of chest compressions with the use of a new audio-visual feedback device: a randomized manikin study in health care professionals.

AIM: Optimal depth (50-60mm) and rate (100-120min(-1)) of chest compressions (CC) is the prerequisite of effective cardiopulmonary resuscitation (CPR). However, insufficient CC during CPR are common even among health care professionals. We sought to evaluate if CC are more effective with the use of a novel feedback device compared to standard CC. Primary endpoints were absolute percentage of correct CC of all CC (correct rate and correct depth, classified as "optimal" CC), and the percentage of CC in target rate and percentage of CC in target depth. METHODS: 63 healthcare professionals performed CC on a manikin with the use of a novel feedback device. The device provides audio-visual information about compression depth and rate. Each participant performed two minutes of CC with and without feedback. Participants were randomized into two groups that performed either CC with feedback first, followed by a trial without feedback, or vice versa. All participants answered a short questionnaire on self-estimation of CC performance. RESULTS: The absolute percentage of optimal compressions of all compressions has increased from 27.9±28.8% to 47.6±33.5% (p<0.001) with use of the device. Furthermore, a significant increase of the percentage of CC in target depth (35.9±30.6% without vs. 54.8±33.5% with the device, p=0.003) and in target rate (70.5±37.7% without vs. 82.7±27.8 with the device, p=0.039) were observed. CONCLUSION: This novel feedback device significantly improved the quality of CC in health care professionals.

Treatment of Nonagenarians With Atrial Fibrillation: Insights From the Berlin Atrial Fibrillation (BAF) Registry.

OBJECTIVES: The objective of this study was to determine course and treatment of atrial fibrillation (AF) in nonagenarians. Incidence of AF increases with age. Due to the demographic change in the industrialized world, an increase of AF in the group of elderly and very elderly is expected in the next decades. However, only few data exist on the clinical relevance of AF in patients aged 89 years or older. DESIGN: Observational, mono-centric registry. SETTING: University hospital. PARTICIPANTS: Of the 11,888 patients included in the Berlin Atrial Fibrillation (BAF) Registry, 279 patients aged 89 years or older were identified. All patients presented to our hospital with AF between January 2001 and December 2014. MEASUREMENTS: AF type, symptoms, comorbidities, CHA2DS2-VASc and HAS-BLED, treatment strategy, and anticoagulant treatment were assessed at baseline. A composite of stroke/transient ischemic attack (TIA), thromboembolic events, major bleeding, and death was the primary endpoint. Stroke/TIA, thromboembolic events and major bleeding, presence of AF, new onset of heart failure and change of NYHA class, and bradyarrhythmia necessitating pacemaker implantation were secondary endpoints. RESULTS: Patients (age 92 ± 2.7 years, range 89-108) presented in EHRA class I in 38.4% of the cases, class II in 49.5%, class III in 10%, and class IV in 2%. Rhythm control was attempted in 37 (13.3%) of the patients. Baseline CHA2DS2-VASc and HAS-BLED were 5.0 ± 1.3 and 3.1 ± 0.9, respectively. Oral anticoagulation (OAC) was initiated in 74 (26.5 %) of the patients. Of all patients, 33 (11.8%) patients died in hospital. Of the remaining patients, 104 were followed over 13.8 ± 17.5 months with 3.5 ± 2.3 visits during follow-up. Rhythm control was attempted in 10 patients (9.6%). OAC was initiated in 37 patients (35.6 %). Fifty-nine (56.7%) patients reached the primary composite endpoint. Stroke/TIA (34.6%) and heart failure (49%) were common. Subgroup analysis revealed no significant differences in any of the endpoints between patients undergoing rhythm versus rate control and between patients under OAC compared with patients without OAC. INR at follow-up and TTR were 1.76 ± 1.0 and 29.5% ± 37.8% in patients receiving VKA. CONCLUSION: In this real-world cohort of very elderly patients with AF, a rhythm control strategy and OAC treatment were chosen only in a minority of the cases. If OAC was initiated, most received VKAs with a poor TTR during follow-up. A high incidence of stroke/TIA was observed in patients with and without OAC. Further data are needed to define optimal treatment of AF in this particular patient group.