Erratum: [Corrigendum] CD271+ stem cell treatment of patients with chronic stroke: A retrospective case series report.

[This corrects the article DOI: 10.3892/etm.2020.8948.].

Leveraging artificial intelligence to advance the understanding of chemical neurotoxicity.

Neurotoxicology is a specialty that aims to understand and explain the impact of chemicals, xenobiotics and physical conditions on nervous system function throughout the life span. Herein, we point to the need for integration of novel translational bioinformatics and chemo-informatics approaches, such as machine learning (ML) and artificial intelligence (AI) to the discipline. Specifically, we advance the notion that AI and ML will be helpful in identifying neurotoxic signatures, provide reliable data in predicting neurotoxicity in the context of genetic variability, and improve the understanding of neurotoxic outcomes associated with exposures to mixtures, to name a few.

Emerging deep learning techniques using magnetic resonance imaging data applied in multiple sclerosis and clinical isolated syndrome patients (Review).

Computer-aided diagnosis systems aim to assist clinicians in the early identification of abnormal signs in order to optimize the interpretation of medical images and increase diagnostic precision. Multiple sclerosis (MS) and clinically isolated syndrome (CIS) are chronic inflammatory, demyelinating diseases affecting the central nervous system. Recent advances in deep learning (DL) techniques have led to novel computational paradigms in MS and CIS imaging designed for automatic segmentation and detection of areas of interest and automatic classification of anatomic structures, as well as optimization of neuroimaging protocols. To this end, there are several publications presenting artificial intelligence-based predictive models aiming to increase diagnostic accuracy and to facilitate optimal clinical management in patients diagnosed with MS and/or CIS. The current study presents a thorough review covering DL techniques that have been applied in MS and CIS during recent years, shedding light on their current advances and limitations.

Current status and future prospects of PET-imaging applications in patients with gastro-entero-pancreatic neuroendocrine tumors (GEP-NETs).

Gastro-entero-pancreatic neuroendocrine tumors (GEP-NETs) represent a heterogeneous group of rare neoplasms with increasing incidence over the last decades. Localization of GEP-NETs and their metastases is a vital component for the implementation of accurate and patient-tailored treatment strategies. Addressing this challenge requires the employment of multidisciplinary imaging approaches, with hybrid positron emission tomography/computed tomography (PET/CT) imaging techniques standing at the forefront of this effort. GEP-NETs exhibit several pathophysiologic characteristics, which can serve as highly specific molecular targets that can be effectively visualized and quantified by means of PET-radiopharmaceuticals, facilitating diagnosis, accurate staging and efficient monitoring of treatment response. Furthermore, the capability for whole-body, in-vivo, non-invasive characterization of the molecular heterogeneity of the disease, provides strong prognostic information, while enabling the selection of patients suitable for precision-based theranostic approaches. The dual tracer (18F-FDG & 68Ga-DOTA-peptides) PET/CT imaging approach is the current optimal diagnostic imaging strategy, since it enables tumor localization, accurate staging, non-invasive whole-body total tumor burden characterization of disease heterogeneity, while providing strong prognostic information and guidance towards treatment strategy. Moreover, 64Cu-DOTATATE has been recently approved by FDA for SSTRs positive NETs, promising substantial diagnostic and logistical benefits. Furthermore, 18F-DOPA offers diagnostic capabilities for serotonin-secreting GEP-NETs which are not characterized by cell-surface over-expression of somatostatin receptors (SSTRs) and cannot be seen on morphological imaging. In addition, PET/CT with agents targeting the expression of glucagon-like peptide-1 receptor (GLP-R1) should be considered in cases of clinical suspicion for insulinomas that cannot be detected by morphological imaging or STTRs PET/CT imaging.

Pituitary Imaging Abnormalities and Related Endocrine Disorders in Erdheim-Chester Disease.

PURPOSE: We examined abnormal pituitary imaging (API) and associated endocrine dysfunction in subjects with ECD. METHODS: A cross-sectional descriptive examination of a natural history cohort study diagnosed with ECD was conducted at a clinical research center. Subjects underwent baseline endocrine tests of anterior and posterior pituitary function and dedicated pituitary gland MRI scans. We determined the frequency of various pituitary imaging abnormalities in ECD and assessed its relationships with age, sex, body mass index (BMI), BRAF V600E status, high sensitivity C-reactive protein (hsCRP), erythrocyte sedimentation rate (ESR), pituitary hormone deficits and number, diabetes insipidus (DI), and panhypopituitarism. RESULTS: Our cohort included 61 subjects with ECD [age (SD): 54.3 (10.9) y, 46 males/15 females]. API was present in 47.5% (29/61) of ECD subjects. Loss of the posterior pituitary bright spot (36.1%) followed by thickened pituitary stalk (24.6%), abnormal enhancement (18.0%), and pituitary atrophy (14.8%) were the most common abnormalities. DI and panhypopituitarism were more frequent in subjects with API without differences in age, sex distribution, hsCRP, ESR, and BRAF V600E status compared to normal pituitary imaging. CONCLUSIONS: We noted a high burden of API and endocrinopathies in ECD. API was highly associated with the presence of panhypopituitarism and DI. Therefore, a thorough assessment of hypothalamic-pituitary integrity should be considered in subjects with ECD.

Cross-Sectional Imaging Useful in Melorheostosis.

Melorheostosis is a rare disease of bone overgrowth that is primarily diagnosed based on imaging studies. Recently, the association of different radiological patterns of the disease with distinct genetic cause was reported. Several case reports have described the radiological findings in patients with melorheostosis. However, the added value of cross-sectional imaging with CT and MRI beyond X-rays has not been investigated. The aim of the current study was to investigate this existing gap in knowledge. Forty patients with melorheostosis seen at the National Institute of Health Clinical Center were included in the study, and all their imaging studies were analyzed. The sequence of interpretation was X-ray followed by CT and then MRI. CT images were extracted from whole-body 18F-sodium fluoride positron emission tomography/CT studies. The information from CT reclassified the initial X-rays based radiological pattern in 13 patients. Additionally, CT comprehensively identified joint involvement and disease extent. In 76% of patients (n = 29) who underwent MRI, additional findings were noted, ranging from soft tissue edema to identification of soft tissue masses and incidental findings. MRI did not provide additional information on skeletal lesions beyond CT scans. However, it revealed the extension of soft tissue ossification into ischiofemoral space in four patients who complained of deep gluteal pain consistent with ischiofemoral impingement syndrome. In addition, MRI revealed soft tissue edema in 20 patients, 9 of whom had bone marrow edema and periosteal edema in the tibias consistent with shin splints. These findings suggest that select patients with melorheostosis should be evaluated with both CT and MRI, particularly patients in whom the distribution of pain does not correlate with the anatomic location of the disease in plain radiographs. © 2021 The Authors. JBMR Plus published by Wiley Periodicals LLC. on behalf of American Society for Bone and Mineral Research.

Advanced clinical imaging for the evaluation of stem cell based therapies.

Introduction: As stem cell treatments reach closer to the clinic, the need for appropriate noninvasive imaging for accurate disease diagnosis, treatment planning, follow-up, and early detection of complications, is constantly rising. Clinical radiology affords an extensive arsenal of advanced imaging techniques, to provide anatomical and functional information on the whole spectrum of stem cell treatments from diagnosis to follow-up.Areas covered: This manuscript aims at providing a critical review of major published studies on the utilization of advanced imaging for stem cell treatments. Uses of magnetic resonance imaging (MRI), computed tomography (CT), ultrasound, and positron emission tomography (PET) are reviewed and interrogated for their applicability to stem cell imaging.Expert opinion: A wide spectrum of imaging methods have been utilized for the evaluation of stem cell therapies. The majority of published techniques are not clinically applicable, using methods exclusively applicable to animals or technology irrelevant to current clinical practice. Harmonization of preclinical methods with clinical reality is necessary for the timely translation of stem cell therapies to the clinic. Methods such as diffusion weighted MRI, hybrid imaging, and contrast-enhanced ultrasound hold great promise and should be routinely incorporated in the evaluation of patients receiving stem cell treatments.

Volumetric Modeling of Adrenal Gland Size in Primary Bilateral Macronodular Adrenocortical Hyperplasia.

CONTEXT: Radiological characterization of adrenal size in primary bilateral macronodular adrenocortical hyperplasia (PBMAH) has not been previously investigated. OBJECTIVE: We hypothesized that volumetric modeling of adrenal gland size may correlate with biochemical disease severity in patients with PBMAH. Secondary analysis of patients with concurrent primary aldosteronism (PA) was performed. DESIGN: A retrospective cross-sectional analysis of 44 patients with PBMAH was conducted from 2000 to 2019. SETTING: Tertiary care clinical research center. PATIENTS: Patients were diagnosed with PBMAH based upon clinical, genetic, radiographic and biochemical characteristics. INTERVENTION: Clinical, biochemical, and genetic data were obtained. Computed tomography scans were used to create volumetric models by manually contouring both adrenal glands in each slice using Vitrea Core Fx v6.3 software (Vital Images, Minnetonka, Minnesota). MAIN OUTCOME AND MEASURES: 17-hydroxycorticosteroids (17-OHS), ARMC5 genetics, and aldosterone-to-renin ratio (ARR) were retrospectively obtained. Pearson test was used for correlation analysis of biochemical data with adrenal volume. RESULTS: A cohort of 44 patients with PBMAH was evaluated, with a mean age (±SD) of 53 ±â€…11.53. Eight patients met the diagnostic criteria for PA, of whom 6 (75%) were Black. In the Black cohort, total adrenal volumes positively correlated with midnight cortisol (R = 0.76, P = 0.028), urinary free cortisol (R = 0.70, P = 0.035), and 17-OHS (R = 0.87, P = 0.0045), with a more pronounced correlation with left adrenal volume alone. 17-OHS concentration positively correlated with total, left, and right adrenal volume in patients harboring pathogenic variants in ARMC5 (R = 0.72, P = 0.018; R = 0.65, P = 0.042; and R = 0.73, P = 0.016, respectively). CONCLUSIONS: Volumetric modeling of adrenal gland size may associate with biochemical severity in patients with PBMAH, with particular utility in Black patients.

Extended perfusion protocol for MS lesion quantification.

This study aims to examine a time-extended dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) protocol and report a comparative study with three different pharmacokinetic (PK) models, for accurate determination of subtle blood-brain barrier (BBB) disruption in patients with multiple sclerosis (MS). This time-extended DCE-MRI perfusion protocol, called Snaps, was applied on 24 active demyelinating lesions of 12 MS patients. Statistical analysis was performed for both protocols through three different PK models. The Snaps protocol achieved triple the window time of perfusion observation by extending the magnetic resonance acquisition time by less than 2 min on average for all patients. In addition, the statistical analysis in terms of adj-R 2 goodness of fit demonstrated that the Snaps protocol outperformed the conventional DCE-MRI protocol by detecting 49% more pixels on average. The exclusive pixels identified from the Snaps protocol lie in the low k trans range, potentially reflecting areas with subtle BBB disruption. Finally, the extended Tofts model was found to have the highest fitting accuracy for both analyzed protocols. The previously proposed time-extended DCE protocol, called Snaps, provides additional temporal perfusion information at the expense of a minimal extension of the conventional DCE acquisition time.

Prevalence of Hypothyroidism in Patients With Erdheim-Chester Disease.

Importance: Erdheim-Chester disease (ECD) is a rare non-Langerhans cell histiocytosis affecting multiple organs and commonly caused by somatic pathogenic variants in BRAF V600E and mitogen-activated protein kinase genes. Clinical features of ECD result from histiocytic involvement of various tissues; while endocrine involvement in ECD occurs frequently, the prevalence of central or primary hypothyroidism has not been thoroughly investigated. Objective: To assess hypothalamus-pituitary-thyroid (HPT) dysfunction in patients with ECD. Design, Setting, and Participants: This cross-sectional study included 61 patients with ECD who were enrolled in a natural history study at a tertiary care center between January 2011 and December 2018. ECD was diagnosed on the basis of clinical, genetic, and histopathological features. Data were analyzed in March 2020. Exposure: Diagnosis of ECD. Main Outcomes and Measures: Main outcome was the prevalence of thyroid dysfunction in adults with ECD compared with community estimates. Patients underwent baseline evaluation with a thyroid function test, including thyrotropin, free thyroxine (fT4), and total thyroxine (T4), and sellar imaging with magnetic resonance imaging or computed tomography scan. The association of HPT dysfunction was assessed for differences in age, sex, body mass index, BRAF V600E status, high sensitivity C-reactive protein level, sellar imaging, and pituitary hormonal dysfunction. Results: A total of 61 patients with ECD (46 [75%] men; mean [SD] age, 54.3 [10.9] years) were evaluated. Seventeen patients (28%) had hypothyroidism requiring levothyroxine therapy. The prevalence of both central and primary hypothyroidism were higher than community estimates (central hypothyroidism: 9.8% vs 0.1%; odds ratio, 109.0; 95% CI, 37.4-260.6; P < .001; primary hypothyroidism: 18.0% vs 4.7%; OR, 4.4; 95% CI, 2.1-8.7; P < .001). Patients with hypothyroidism (both primary and central), compared with patients with euthyroidism, had higher body mass index (median [interquartile range] 31.4 [28.3-38.3] vs 26.7 [24.4-31.9]; P = .004) and a higher prevalence of panhypopituitarism (7 [47%] vs 3 [7%]; P < .001). Among patients with hypothyroidism, those with central hypothyroidism, compared with patients with primary hypothyroidism, had a lower mean (SD) body mass index (28.3 [2.6] vs 36.3 [5.9]; P = .007) and higher frequencies of abnormal sellar imaging (5 [83%] vs 3 [27%]; P = .050) and panhypopituitarism (5 [83%] vs 3 [27%]; P = .050). Conclusions and Relevance: In this cohort study, a higher prevalence of central and primary hypothyroidism was identified in patients with ECD compared with the community. There should be a low threshold for testing for hypothyroidism in patients with ECD, and treatment should follow standard guidelines.

Diffusion Tensor Imaging and Chemical Exchange Saturation Transfer MRI Evaluation on the Long-Term Effects of Pulsed Focused Ultrasound and Microbubbles Blood Brain Barrier Opening in the Rat.

Blood-brain barrier opening (BBBO) with pulsed Focused Ultrasound (pFUS) and microbubbles (MB) has received increasing interest as a method for neurotherapeutics of the central nervous system. In general, conventional MRI [i.e., T2w, T2∗w, gadolinium (Gd) enhanced T1w] is used to monitor the effects of pFUS+MB on BBBO and/or assess whether sonication results in parenchymal damage. This study employed multimodal MRI techniques and 18F-Fludeoxyglucose (FDG) PET to evaluate the effects of single and multiple weekly pFUS+MB sessions on morphology and glucose utilization levels in the rat cortex and hippocampus. pFUS was performed with 0.548 MHz transducer with a slow infusion over 1 min of OptisonTM (5-8 × 107 MB) in nine focal points in cortex and four in hippocampus. During pFUS+MB treatment, Gd-T1w was performed at 3 T to confirm BBBO, along with subsequent T2w, T2∗w, DTI and glucose CEST (glucoCEST)-weighted imaging by high field 9.4 T and compared with FDG-PET and immunohistochemistry. Animals receiving a single pFUS+MB exhibited minimal hypointense voxels on T2∗w. Brains receiving multiple pFUS+MB treatments demonstrated persistent T2w and T2∗ abnormalities associated with changes in DTI and glucoCEST when compared to contralateral parenchyma. Decreased glucoCEST contrast was substantiated by FDG-PET in cortex following multiple sonications. Immunohistochemistry showed significantly dilated vessels and decreased neuronal glucose transporter (GLUT3) expression in sonicated cortex and hippocampus without changes in neuronal counts. These results suggest the importance to standardize MRI protocols in concert with advanced imaging techniques when evaluating long term effects of pFUS+MB BBBO in clinical trials for neurological diseases.

Musculoskeletal trauma imaging in the era of novel molecular methods and artificial intelligence.

Over the past decade rapid advancements in molecular imaging (MI) and artificial intelligence (AI) have revolutionized traditional musculoskeletal radiology. Molecular imaging refers to the ability of various methods to in vivo characterize and quantify biological processes, at a molecular level. The extracted information provides the tools to understand the pathophysiology of diseases and thus to early detect, to accurately evaluate the extend and to apply and evaluate targeted treatments. At present, molecular imaging mainly involves CT, MRI, radionuclide, US, and optical imaging and has been reported in many clinical and preclinical studies. Although originally MI techniques targeted at central nervous system disorders, later on their value on musculoskeletal disorders was also studied in depth. Meaningful exploitation of the large volume of imaging data generated by molecular and conventional imaging techniques, requires state-of-the-art computational methods that enable rapid handling of large volumes of information. AI allows end-to-end training of computer algorithms to perform tasks encountered in everyday clinical practice including diagnosis, disease severity classification and image optimization. Notably, the development of deep learning algorithms has offered novel methods that enable intelligent processing of large imaging datasets in an attempt to automate decision-making in a wide variety of settings related to musculoskeletal trauma. Current applications of AI include the diagnosis of bone and soft tissue injuries, monitoring of the healing process and prediction of injuries in the professional sports setting. This review presents the current applications of novel MI techniques and methods and the emerging role of AI regarding the diagnosis and evaluation of musculoskeletal trauma.

Advancing COVID-19 differentiation with a robust preprocessing and integration of multi-institutional open-repository computer tomography datasets for deep learning analysis.

The coronavirus pandemic and its unprecedented consequences globally has spurred the interest of the artificial intelligence research community. A plethora of published studies have investigated the role of imaging such as chest X-rays and computer tomography in coronavirus disease 2019 (COVID-19) automated diagnosis. Οpen repositories of medical imaging data can play a significant role by promoting cooperation among institutes in a world-wide scale. However, they may induce limitations related to variable data quality and intrinsic differences due to the wide variety of scanner vendors and imaging parameters. In this study, a state-of-the-art custom U-Net model is presented with a dice similarity coefficient performance of 99.6% along with a transfer learning VGG-19 based model for COVID-19 versus pneumonia differentiation exhibiting an area under curve of 96.1%. The above was significantly improved over the baseline model trained with no segmentation in selected tomographic slices of the same dataset. The presented study highlights the importance of a robust preprocessing protocol for image analysis within a heterogeneous imaging dataset and assesses the potential diagnostic value of the presented COVID-19 model by comparing its performance to the state of the art.

CD271+ stem cell treatment of patients with chronic stroke: : A retrospective case series report.

Patients with chronic stroke have currently little hope for motor improvement towards regaining independent activities of daily living; stem cell treatments offer a new treatment option and needs to be developed. Patients with chronic stroke (more than 3 months prior to stem cell treatment, mean 21.2 months post-stroke) were treated with CD271+ stem cells, 7 patients received autologous and 1 allogeneic cells from first degree relative; administration was intravenous in 1 and intrathecal in 7 patients. Each patient received a single treatment consisting of 2-5x106 cells/kg and they were followed up for up to 12 months. There were significant improvements in expressive aphasia (2/3 patients) spasticity (5/5, of which 2 were transient), and small improvements in motor function (2/8 patients). Although motor improvements were minor in our chronic stroke patients, improvements in aphasia and spasticity were significant and in the context of good safety we are advocating further administration and clinical studies of CD271+ stem cells not only in chronic stroke patients, but also for spastic paresis/plegia; a different, yet unexplored application is pulmonary emphysema.

Interpretable artificial intelligence framework for COVID-19 screening on chest X-rays.

COVID-19 has led to an unprecedented healthcare crisis with millions of infected people across the globe often pushing infrastructures, healthcare workers and entire economies beyond their limits. The scarcity of testing kits, even in developed countries, has led to extensive research efforts towards alternative solutions with high sensitivity. Chest radiological imaging paired with artificial intelligence (AI) can offer significant advantages in diagnosis of novel coronavirus infected patients. To this end, transfer learning techniques are used for overcoming the limitations emanating from the lack of relevant big datasets, enabling specialized models to converge on limited data, as in the case of X-rays of COVID-19 patients. In this study, we present an interpretable AI framework assessed by expert radiologists on the basis on how well the attention maps focus on the diagnostically-relevant image regions. The proposed transfer learning methodology achieves an overall area under the curve of 1 for a binary classification problem across a 5-fold training/testing dataset.

A dissection of SARS­CoV2 with clinical implications (Review).

We are being confronted with the most consequential pandemic since the Spanish flu of 1918­1920 to the extent that never before have 4 billion people quarantined simultaneously; to address this global challenge we bring to the forefront the options for medical treatment and summarize SARS­CoV2 structure and functions, immune responses and known treatments. Based on literature and our own experience we propose new interventions, including the use of amiodarone, simvastatin, pioglitazone and curcumin. In mild infections (sore throat, cough) we advocate prompt local treatment for the naso­pharynx (inhalations; aerosols; nebulizers); for moderate to severe infections we propose a tried­and­true treatment: the combination of arginine and ascorbate, administered orally or intravenously. The material is organized in three sections: i) Clinical aspects of COVID­19; acute respiratory distress syndrome (ARDS); known treatments; ii) Structure and functions of SARS­CoV2 and proposed antiviral drugs; iii) The combination of arginine­ascorbate.

Artificial intelligence radiogenomics for advancing precision and effectiveness in oncologic care (Review).

The new era of artificial intelligence (AI) has introduced revolutionary data­driven analysis paradigms that have led to significant advancements in information processing techniques in the context of clinical decision­support systems. These advances have created unprecedented momentum in computational medical imaging applications and have given rise to new precision medicine research areas. Radiogenomics is a novel research field focusing on establishing associations between radiological features and genomic or molecular expression in order to shed light on the underlying disease mechanisms and enhance diagnostic procedures towards personalized medicine. The aim of the current review was to elucidate recent advances in radiogenomics research, focusing on deep learning with emphasis on radiology and oncology applications. The main deep learning radiogenomics architectures, together with the clinical questions addressed, and the achieved genetic or molecular correlations are presented, while a performance comparison of the proposed methodologies is conducted. Finally, current limitations, potentially understudied topics and future research directions are discussed.