Cognitive trajectories in multiple sclerosis: a long-term follow-up study.

BACKGROUND: Cognitive impairment occurs in multiple sclerosis (MS) and undergoes a progressive worsening over disease course. However, clinicians still struggle to predict the course of cognitive function. To evaluate baseline clinical and imaging predictors of cognitive abilities worsening over time, we performed a latent trajectory analysis for cognitive performances in MS patients, up to 15 years from disease onset. METHODS: We collected age, sex, education, dominant and non-dominant 9-hole peg test (9HP) and timed 25-foot walk (T25-FW) as well as MRI measures (grey matter volume and lesion load) within 6 months from disease diagnosis for relapsing-remitting MS (RR-MS) patients. At diagnosis and over the follow-up, we also assessed cognitive status through the symbol digit modalities test (SDMT). Cognitive impairment was defined by applying age-, gender- and education-adjusted normative values. Group-based trajectory analysis was performed to determine trajectories, and the predictive value of clinical and imaging variables at baseline was assessed through multinomial logistic regression. RESULTS: We included 148 RR-MS (98 females and 50 males). Over 11 ± 4 year follow-up, 51.4% remained cognitively stable whereas 48.6% cognitively worsened. Cognitively worsening patients had a higher T25FW time (p = 0.004) and a reduced hippocampal volume at baseline (p = 0.04). CONCLUSION: Physical disability as well as hippocampal atrophy might depict patients at risk of cognitive worsening over the disease course. Therefore, using such predictors, clinicians may select patients to carefully evaluate for cognitive impairment as to eventually introduce cognitive rehabilitation treatments.

Clinical report of a brain magnetic resonance imaging finding in Noonan syndrome.

Noonan syndrome (NS) is an autosomal dominant disease caused by aberrant up-regulated signaling through RAS GTPase. It is characterized by facial dysmorphisms, short stature, congenital heart defects, malformations of rib cage bones, bleeding problems, learning difficulties, or mild intellectual disability. Additional intracranial findings in NS patients include tumors, midline anomalies, and malformations of cortical development. In this report, we present the case of a young female patient, with a known diagnosis of Noonan syndrome that in complete well being developed two brain lesions, in the right nucleus pallidus and in the left cerebellar hemisphere respectively, whose location and signal on MRI looked similar to neurofibromatosis type 1 unidentified bright objects (UBOs), and whose spectroscopic characteristics excluded neoplasms.

Machine learning for the identification of clinically significant prostate cancer on MRI: a meta-analysis.

OBJECTIVES: The aim of this study was to systematically review the literature and perform a meta-analysis of machine learning (ML) diagnostic accuracy studies focused on clinically significant prostate cancer (csPCa) identification on MRI. METHODS: Multiple medical databases were systematically searched for studies on ML applications in csPCa identification up to July 31, 2019. Two reviewers screened all papers independently for eligibility. The area under the receiver operating characteristic curves (AUC) was pooled to quantify predictive accuracy. A random-effects model estimated overall effect size while statistical heterogeneity was assessed with the I2 value. A funnel plot was used to investigate publication bias. Subgroup analyses were performed based on reference standard (biopsy or radical prostatectomy) and ML type (deep and non-deep). RESULTS: After the final revision, 12 studies were included in the analysis. Statistical heterogeneity was high both in overall and in subgroup analyses. The overall pooled AUC for ML in csPCa identification was 0.86, with 0.81-0.91 95% confidence intervals (95%CI). The biopsy subgroup (n = 9) had a pooled AUC of 0.85 (95%CI = 0.79-0.91) while the radical prostatectomy one (n = 3) of 0.88 (95%CI = 0.76-0.99). Deep learning ML (n = 4) had a 0.78 AUC (95%CI = 0.69-0.86) while the remaining 8 had AUC = 0.90 (95%CI = 0.85-0.94). CONCLUSIONS: ML pipelines using prostate MRI to identify csPCa showed good accuracy and should be further investigated, possibly with better standardisation in design and reporting of results. KEY POINTS: • Overall pooled AUC was 0.86 with 0.81-0.91 95% confidence intervals. • In the reference standard subgroup analysis, algorithm accuracy was similar with pooled AUCs of 0.85 (0.79-0.91 95% confidence intervals) and 0.88 (0.76-0.99 95% confidence intervals) for studies employing biopsies and radical prostatectomy, respectively. • Deep learning pipelines performed worse (AUC = 0.78, 0.69-0.86 95% confidence intervals) than other approaches (AUC = 0.90, 0.85-0.94 95% confidence intervals).

2D linear measures of ventricular enlargement may be relevant markers of brain atrophy and long-term disability progression in multiple sclerosis.

OBJECTIVES: Aim of this study was to investigate the reliability and validity of 2D linear measures of ventricular enlargement as indirect markers of brain atrophy and possible predictors of clinical disability. METHODS: In this retrospective longitudinal analysis of relapsing-remitting MS patients, brain volumes were computed at baseline and after 2 years. Frontal horn width (FHW), intercaudate distance (ICD), third ventricle width (TVW), and 4th ventricle width were obtained. Two-dimensional measures associated with brain volume at correlation analyses were entered in linear and logistic regression models testing the relationship with baseline clinical disability and 10-year confirmed disability progression (CDP), respectively. Possible cutoff values for clinically relevant atrophy were estimated via receiver operating characteristic (ROC) analyses and probed as 10-year CDP predictors using hierarchical logistic regression. RESULTS: Eighty-seven patients were available (61/26 = F/M; 34.1 ± 8.5 years). Moderate negative correlations emerged between ICD and TVW and normalized brain volume (NBV; p < 0.001) and percentage brain volume change per year (PBVC/y) and FHW, ICD, and TVW annual changes (p ≤ 0.005). Baseline disability was moderately associated with NBV, ICD, and TVW (p < 0.001), while PBVC/y predicted 10-year CDP (p = 0.01). A cutoff percentage ICD change per year (PICDC/y) value of 4.38%, corresponding to - 0.91% PBVC/y, correlated with 10-year CDP (p = 0.04). These estimated cutoff values provided extra value for predicting 10-year CDP (PBVC/y: p = 0.001; PICDC/y: p = 0.03). CONCLUSIONS: Two-dimensional measures of ventricular enlargement are reproducible and clinically relevant markers of brain atrophy, with ICD and its increase over time showing the best association with clinical disability. Specifically, a cutoff PICDC/y value of 4.38% could serve as a potential surrogate marker of long-term disability progression. KEY POINTS: • Assessment of ventricular enlargement as a rapidly accessible indirect marker of brain atrophy may prove useful in cases in which brain volume quantification is not practicable. • Two-dimensional linear measures of ventricular enlargement represent reliable, valid, and clinically relevant markers of brain atrophy. • A cutoff annualized percentage brain volume change of - 0.91% and the corresponding annualized percentage increase of 4.38% for intercaudate distance are able to discriminate patients who will develop long-term disability progression.

Giant Basal Cell Carcinoma of the Scalp with Intracranial Invasion: MRI Findings with Tract Visualisation.

A rare case of recurrent basal cell carcinoma in the scalp that infiltrated multiple intracranial structures is presented. Basal cell carcinoma represents one of the most frequent malignant nonmelanotic skin neoplasms, but the majority of them have no aggressive and recurrent behaviour. The aim of this case report is to provide an overview of the main clinical and radiologic features of basal cell carcinoma, focusing on the conventional and advanced (tractography) MRI findings and providing an overview of treatment and prognosis.

Machine learning applications in prostate cancer magnetic resonance imaging.

With this review, we aimed to provide a synopsis of recently proposed applications of machine learning (ML) in radiology focusing on prostate magnetic resonance imaging (MRI). After defining the difference between ML and classical rule-based algorithms and the distinction among supervised, unsupervised and reinforcement learning, we explain the characteristic of deep learning (DL), a particular new type of ML, including its structure mimicking human neural networks and its 'black box' nature. Differences in the pipeline for applying ML and DL to prostate MRI are highlighted. The following potential clinical applications in different settings are outlined, many of them based only on MRI-unenhanced sequences: gland segmentation; assessment of lesion aggressiveness to distinguish between clinically significant and indolent cancers, allowing for active surveillance; cancer detection/diagnosis and localisation (transition versus peripheral zone, use of prostate imaging reporting and data system (PI-RADS) version 2), reading reproducibility, differentiation of cancers from prostatitis benign hyperplasia; local staging and pre-treatment assessment (detection of extraprostatic disease extension, planning of radiation therapy); and prediction of biochemical recurrence. Results are promising, but clinical applicability still requires more robust validation across scanner vendors, field strengths and institutions.