Assessing brain involvement in Fabry disease with deep learning and the brain-age paradigm

While neurological manifestations are core features of Fabry disease (FD), quantitative neuroimaging biomarkers allowing to measure brain involvement are lacking. We used deep learning and the brain-age paradigm to assess whether FD patients' brains appear older than normal and to validate brain-predicted age difference (brain-PAD) as a possible disease severity biomarker. MRI scans of FD patients and healthy controls (HCs) from a single Institution were, retrospectively, studied. The Fabry stabilization index (FASTEX) was recorded as a measure of disease severity. Using minimally preprocessed 3D T1-weighted brain scans of healthy subjects from eight publicly available sources (N = 2160; mean age = 33 years [range 4-86]), we trained a model predicting chronological age based on a DenseNet architecture and used it to generate brain-age predictions in the internal cohort. Within a linear modeling framework, brain-PAD was tested for age/sex-adjusted associations with diagnostic group (FD vs. HC), FASTEX score, and both global and voxel-level neuroimaging measures. We studied 52 FD patients (40.6 ± 12.6 years; 28F) and 58 HC (38.4 ± 13.4 years; 28F). The brain-age model achieved accurate out-of-sample performance (mean absolute error = 4.01 years, R2 = .90). FD patients had significantly higher brain-PAD than HC (estimated marginal means: 3.1 vs. -0.1, p = .01). Brain-PAD was associated with FASTEX score (B = 0.10, p = .02), brain parenchymal fraction (B = -153.50, p = .001), white matter hyperintensities load (B = 0.85, p = .01), and tissue volume reduction throughout the brain. We demonstrated that FD patients' brains appear older than normal. Brain-PAD correlates with FD-related multi-organ damage and is influenced by both global brain volume and white matter hyperintensities, offering a comprehensive biomarker of (neurological) disease severity.

Multi input-Multi output 3D CNN for dementia severity assessment with incomplete multimodal data

Alzheimer's Disease is the most common cause of dementia, whose progression spans in different stages, from very mild cognitive impairment to mild and severe conditions. In clinical trials, Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) are mostly used for the early diagnosis of neurodegenerative disorders since they provide volumetric and metabolic function information of the brain, respectively. In recent years, Deep Learning (DL) has been employed in medical imaging with promising results. Moreover, the use of the deep neural networks, especially Convolutional Neural Networks (CNNs), has also enabled the development of DL-based solutions in domains characterized by the need of leveraging information coming from multiple data sources, raising the Multimodal Deep Learning (MDL). In this paper, we conduct a systematic analysis of MDL approaches for dementia severity assessment exploiting MRI and PET scans. We propose a Multi Input-Multi Output 3D CNN whose training iterations change according to the characteristic of the input as it is able to handle incomplete acquisitions, in which one image modality is missed. Experiments performed on OASIS-3 dataset show the satisfactory results of the implemented network, which outperforms approaches exploiting both single image modality and different MDL fusion techniques.

3T-MRI Artificial Intelligence in Patients with Invasive Breast Cancer to Predict Distant Metastasis Status: A Pilot Study

BACKGROUND

The incidence of breast cancer metastasis has decreased over the years. However, 20-30% of patients with early breast cancer still die from metastases. The purpose of this study is to evaluate the performance of a Deep Learning Convolutional Neural Networks (CNN) model to predict the risk of distant metastasis using 3T-MRI DCE sequences (Dynamic Contrast-Enhanced).

METHODS

A total of 157 breast cancer patients who underwent staging 3T-MRI examinations from January 2011 to July 2022 were retrospectively examined. Patient data, tumor histological and MRI characteristics, and clinical and imaging follow-up examinations of up to 7 years were collected. Of the 157 MRI examinations, 39/157 patients (40 lesions) had distant metastases, while 118/157 patients (120 lesions) were negative for distant metastases (control group). We analyzed the role of the Deep Learning technique using a single variable size bounding box (SVB) option and employed a Voxel Based (VB) NET CNN model. The CNN performance was evaluated in terms of accuracy, sensitivity, specificity, and area under the ROC curve (AUC).

RESULTS

The VB-NET model obtained a sensitivity, specificity, accuracy, and AUC of 52.50%, 80.51%, 73.42%, and 68.56%, respectively. A significant correlation was found between the risk of distant metastasis and tumor size, and the expression of PgR and HER2.

CONCLUSIONS

We demonstrated a currently insufficient ability of the Deep Learning approach in predicting a distant metastasis status in patients with BC using CNNs.

CNN-Based Approaches with Different Tumor Bounding Options for Lymph Node Status Prediction in Breast DCE-MRI

Simple Summary

Breast cancer represents the most frequent cancer in women in the world. The state of the axillary lymph node is considered an independent prognostic factor and is currently evaluated only with invasive methods. Deep learning approaches, especially the ones based on convolutional neural networks, offer a valid, non-invasive alternative, allowing extraction of large amounts of the quantitative data that are used to build predictive models. The aim of our work is to evaluate the influence of the peritumoral parenchyma through different bounding box techniques on the prediction of the axillary lymph node in breast cancer patients using a deep learning artificial intelligence approach.

Abstract

Background: The axillary lymph node status (ALNS) is one of the most important prognostic factors in breast cancer (BC) patients, and it is currently evaluated by invasive procedures. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), highlights the physiological and morphological characteristics of primary tumor tissue. Deep learning approaches (DL), such as convolutional neural networks (CNNs), are able to autonomously learn the set of features directly from images for a specific task. Materials and Methods: A total of 155 malignant BC lesions evaluated via DCE-MRI were included in the study. For each patient’s clinical data, the tumor histological and MRI characteristics and axillary lymph node status (ALNS) were assessed. LNS was considered to be the final label and dichotomized (LN+ (27 patients) vs. LN− (128 patients)). Based on the concept that peritumoral tissue contains valuable information about tumor aggressiveness, in this work, we analyze the contributions of six different tumor bounding options to predict the LNS using a CNN. These bounding boxes include a single fixed-size box (SFB), a single variable-size box (SVB), a single isotropic-size box (SIB), a single lesion variable-size box (SLVB), a single lesion isotropic-size box (SLIB), and a two-dimensional slice (2DS) option. According to the characteristics of the volumes considered as inputs, three different CNNs were investigated: the SFB-NET (for the SFB), the VB-NET (for the SVB, SIB, SLVB, and SLIB), and the 2DS-NET (for the 2DS). All the experiments were run in 10-fold cross-validation. The performance of each CNN was evaluated in terms of accuracy, sensitivity, specificity, the area under the ROC curve (AUC), and Cohen’s kappa coefficient (K). Results: The best accuracy and AUC are obtained by the 2DS-NET (78.63% and 77.86%, respectively). The 2DS-NET also showed the highest specificity, whilst the highest sensibility was attained by the VB-NET based on the SVB and SIB as bounding options. Conclusion: We have demonstrated that a selective inclusion of the DCE-MRI’s peritumoral tissue increases accuracy in the lymph node status prediction in BC patients using CNNs as a DL approach.

Stratification of multiple sclerosis patients using unsupervised machine learning: a single-visit MRI-driven approach

OBJECTIVES

To stratify patients with multiple sclerosis (pwMS) based on brain MRI-derived volumetric features using unsupervised machine learning.

METHODS

The 3-T brain MRIs of relapsing-remitting pwMS including 3D-T1w and FLAIR-T2w sequences were retrospectively collected, along with Expanded Disability Status Scale (EDSS) scores and long-term (10 ± 2 years) clinical outcomes (EDSS, cognition, and progressive course). From the MRIs, volumes of demyelinating lesions and 116 atlas-defined gray matter regions were automatically segmented and expressed as z-scores referenced to external populations. Following feature selection, baseline MRI-derived biomarkers entered the Subtype and Stage Inference (SuStaIn) algorithm, which estimates subgroups characterized by distinct patterns of biomarker evolution and stages within subgroups. The trained model was then applied to longitudinal MRIs. Stability of subtypes and stage change over time were assessed via Krippendorf's α and multilevel linear regression models, respectively. The prognostic relevance of SuStaIn classification was assessed with ordinal/logistic regression analyses.

RESULTS

We selected 425 pwMS (35.9 ± 9.9 years; F/M: 301/124), corresponding to 1129 MRI scans, along with healthy controls (N = 148; 35.9 ± 13.0 years; F/M: 77/71) and external pwMS (N = 80; 40.4 ± 11.9 years; F/M: 56/24) as reference populations. Based on 11 biomarkers surviving feature selection, two subtypes were identified, designated as "deep gray matter (DGM)-first" subtype (N = 238) and "cortex-first" subtype (N = 187) according to the atrophy pattern. Subtypes were consistent over time (α = 0.806), with significant annual stage increase (b = 0.20; p < 0.001). EDSS was associated with stage and DGM-first subtype (p ≤ 0.02). Baseline stage predicted long-term disability, transition to progressive course, and cognitive impairment (p ≤ 0.03), with the latter also associated with DGM-first subtype (p = 0.005).

CONCLUSIONS

Unsupervised learning modelling of brain MRI-derived volumetric features provides a biologically reliable and prognostically meaningful stratification of pwMS.

KEY POINTS

• The unsupervised modelling of brain MRI-derived volumetric features can provide a single-visit stratification of multiple sclerosis patients. • The so-obtained classification tends to be consistent over time and captures disease-related brain damage progression, supporting the biological reliability of the model. • Baseline stratification predicts long-term clinical disability, cognition, and transition to secondary progressive course.

Plasmon Modes of Graphene Nanoribbons with Periodic Planar Arrangements

Among their amazing properties, graphene and related low-dimensional materials show quantized charge-density fluctuations-known as plasmons-when exposed to photons or electrons of suitable energies. Graphene nanoribbons offer an enhanced tunability of these resonant modes, due to their geometrically controllable band gaps. The formidable effort made over recent years in developing graphene-based technologies is however weakened by a lack of predictive modeling approaches that draw upon available ab initio methods. An example of such a framework is presented here, focusing on narrow-width graphene nanoribbons, organized in periodic planar arrays. Time-dependent density-functional calculations reveal unprecedented plasmon modes of different nature at visible to infrared energies. Specifically, semimetallic (zigzag) nanoribbons display an intraband plasmon following the energy-momentum dispersion of a two-dimensional electron gas. Semiconducting (armchair) nanoribbons are instead characterized by two distinct intraband and interband plasmons, whose fascinating interplay is extremely responsive to either injection of charge carriers or increase in electronic temperature. These oscillations share some common trends with recent nanoinfrared imaging of confined edge and surface plasmon modes detected in graphene nanoribbons of 100-500 nm width.

Increased left atrial size in obese children and its association with insulin resistance: a pilot study

Subclinical cardiac abnormalities represent predisposing factors for cardiovascular disease (CVD) in obese subjects. The aim of this study was to evaluate early cardiac abnormalities in obese youth and the potential association with insulin resistance (IR). Thirty obese (12 males (M)/18 females (F); age = 11.5 ± 2.4 years; body mass index (BMI)-standard deviation score (SDS) = +2.1 ± 0.5) and 15 normal weight (10 M/5 F; age = 12.8 ± 3.1 years; BMI-SDS = +0.3 ± 0.9) children and adolescents underwent Doppler two-dimensional echocardiographic assessments of left atrial (LA) and ventricular (LV) geometry and LV diastolic function (peak early [E] and late waves, E wave deceleration time, myocardial flow velocities). Homeostasis model assessment of IR (HOMA-IR) was used as an IR index. LA size was increased in obese children, as indicated by higher LA diameter (4.9 ± 0.5 vs 4.1 ± 0.4 cm, p < 0.001), area (14.3 ± 2.5 vs 10.7 ± 2.0 cm(2), p < 0.001), and volume (33.8 ± 10.6 vs 23.7 ± 6.4 ml, p = 0.003). LV mass was also increased in obese children (87.0 ± 16.6 vs 68.8 ± 13.2 g, p = 0.003), who also showed subtle diastolic dysfunctions, as indicated by higher values of E (97.1 ± 14.3 vs 86.2 ± 11.9 cm/s, p = 0.02). All the above parameters were significantly associated with BMI-SDS (p < 0.05). In addition, HOMA-IR was independently associated with LA diameter, area, and volume (β = 0.314, p = 0.040; β = 0.415, p = 0.008; β = 0.535, p = 0.001).

CONCLUSION

Obese children feature increased LA size, which emerged to be mainly correlated to, and possibly driven by IR, suggesting an increased CVD risk.

WHAT IS KNOWN

Left atrial and ventricular alterations have been reported in obese adults, and they represent predisposing factors for cardiovascular disease. There is some evidence suggesting that obese children show increased left ventricular mass and also increased atrial size, although with conflicting results.

WHAT IS NEW

Obese normotensive children showed a moderately increased atrial size, subtle alterations in left cardiac diastolic function, and ventricular mass. An association between insulin resistance and left cardiac changes was found, although its mechanism remains to be determined.

Phase transitions in strongly coupled three-dimensional Z(N) lattice gauge theories at finite temperature

We perform an analytical and numerical study of the phase transitions in three-dimensional Z(N) lattice gauge theories at finite temperature for N>4, exploiting equivalence of these models with a generalized version of the two-dimensional vector Potts models in the limit of vanishing spatial coupling. In this limit the Polyakov loops play the role of Z(N) spins. The effective couplings of these two-dimensional spin models are calculated explicitly. It is argued that the effective spin models have two phase transitions of BKT type. This is confirmed by large-scale Monte Carlo simulations. Using a cluster algorithm we locate the position of the critical points and study the critical behavior across both phase transitions in details. In particular, we determine various critical indices and compute the helicity modulus, the average action, and the specific heat. A scaling formula for the critical points with N is proposed.

Phase transitions in two-dimensional Z(N) vector models for N>4

We investigate both analytically and numerically the renormalization group equations in two-dimensional (2D) Z(N) vector models. The position of the critical points of the two phase transitions for N>4 is established and the critical index ν is computed. For N=7 and 17 the critical points are located by Monte Carlo simulations, and some of the corresponding critical indices are determined. The behavior of the helicity modulus is studied for N=5, 7, and 17. Using these and other available Monte Carlo data we discuss the scaling of the critical points with N and some other open theoretical problems.

Parasellar myxoid chondrosarcoma: a rare variant of cranial chondrosarcoma

Primary cranial chondrosarcoma in an uncommon cartilaginous tumours of which the myxoid variant is the least reported in the literature. This tumour accounts for 0.15% of all primary intracranial lesions and 6% of all skull base tumours. Chondrosarcomas are frequently misdiagnosed as chordomas, which have a different prognosis. Differential diagnosis is very important because, when treated with similar aggressive treatment strategies, chondrosarcoma has a much better prognosis than chordoma. We describe a 54-year-old female with a 9-month history of left ophtalmoplegia and increasing headache. MR imaging of the head showed a sellar and left parasellar mass. We performed a gross total removal of the mass via a left pterional approach. The histopathologic diagnosis was of a myxoid chondrosarcoma. A post-operative contrast-enhanced computed tomography (CT) scan of the head showed a total removal of the neoplasm. After surgery, the patient showed a transitory dysphasia with right hemiparesis, but they both considerably improved before discharge. Review of the literature was identified using the Medline database: only 10 cases in the worldwide literature were identified to report on this kind of tumour. We present a case report of myxoid chondrosarcoma, a rare variant of chondrosarcoma, with sellar and left parasellar localization. This histopathological type is a low-grade variant and its total removal is effective.

Spontaneous chronic subdural hematoma in a young adult

Chronic subdural hematoma (CSDH) is a common pathology in the elderly but very rare in young adults. When CSDH occurs in this age group, severe head injury or some promotive factors are usually present. This article reports the case of a 29-year-old female presented at our Emergency Department with a few days' history of progressive frontal headache. Computed tomography scan of the head showed a right frontal CSDH. Only a decreased level of consciousness without focal deficits was present at clinical examination and her medical history was negative for trauma or promotive factors. Blood count showed a mild sideropenic anemia while coagulation tests were normal. No vascular malformations were shown at digital subtraction angiography. The patient underwent craniotomic evacuation. After surgery, the patient showed a full neurological recovery. Spontaneous CSDH in young adults is very rare. In the worldwide literature, many cases of non-traumatic CSDH are reported, but a promotive factor is generally present. We described a case of spontaneous CSDH, whose etiology remains unknown.

[Elevated levels of lipoprotein(a) are present in subjects with early ischemic cardiopathy and with a familial history of ischemic cardiopathy]

BACKGROUND

Elevated levels of lipoprotein(a) are associated with a greater risk of atherothrombotic cardiovascular diseases. Since the Lp(a) levels are genetically determined and fairly stable in the course of life and a family history appears to be an independent risk factor of cardiovascular diseases, we evaluated the behavior of Lp(a) levels in patients with early events of coronary heart disease (CHD) and also in subjects with positive family history of ischemic heart diseases.

METHODS

The levels of lipoprotein (a) [Lp(a)] were measured in 254 subjects, 138 males and 116 females with an average age of 48.6 +/- 13.8 years (range 20-76 years). Diabetic subjects, females submitted to oestrogen treatment and those already in treatment with hypolipidaemic drugs were excluded from the study. Forty of the 254 patients (15.7%), 27 males and 13 females, had CHD (29 a previous myocardial infarction and 11 a stable angina). A positive family history for CHD was considered present (102 of the 254 patients) if one or more first degree relatives had angina or myocardial infarction before the age of 60 years in men and 65 in women.

RESULTS

The levels of Lp(a) were higher (p < 0.01) in women (25.1 +/- 28.3 mg/dl) compared to men (17.6 +/- 18.4 mg/dl), without differences in relation to age. The Lp(a) plasmatic levels were not correlated with age, body mass index, total cholesterol, LDL and HDL, triglycerides, apo B, apo AI, fibrinogen and there were no differences in Lp(a) levels in presence or absence of other known cardiovascular risk factors such as hypertension and smoking. The Lp(a) levels were not different between subjects with CHD (28.15 +/- 31.7 mg/dl) and controls (20.3 +/- 22.8 mg/dl). The subjects with CHD were older and had higher levels of fibrinogen and a significantly greater prevalence of hypertension and family history of CHD. Fifteen of the 40 subjects with CHD had an early onset of CHD (before 50 years of age) and only in such patients the Lp(a) levels were significantly greater compared to controls (35.8 +/- 33.2 mg/dl vs 20.3 +/- 22.8 of the controls, p < 0.01), independently of other variables (age, BMI, smoking, hypertension, cholesterol, triglycerides, HDL-c, LDL-c, fibrinogen). Furthermore the Lp(a) plasmatic levels were higher in subjects with a family history of CHD (28.3 +/- 27.6 mg/dl vs 16.3 +/- 18.6 mg/dl of the subjects without a family history of CHD, p < 0.01) even if they had or not had a previous coronary ischemic event.

CONCLUSIONS

Such data confirm the importance of high levels of Lp(a) above all for the early events of CHD and for the subjects with a family history of CHD, which could be expression of a greater predisposition for cardiovascular events.