Faculty Bibliography

The goal of the MRI4ALL hackathon, which took place in October 2023, was to develop a functional low-field MRI scanner in just one week and to release all created source code and resources as open-source packages. The event was attended by 52 participants from 16 institutions who assembled the scanner on the last day of the hackathon. The scanner's magnetic B₀ field with a strength of 43 mT and a target field-of-view size of 11 cm³ was created with a Halbach array made from 990 N40UH permanent magnets, held in place using 3D printed ring formers. Gradient coils were fabricated by gluing enameled copper wire onto 3D printed holders with imprinted wire patterns. A solenoid coil for RF transmission and reception was built by winding 20 turns of Litz wire around a 3D printed cylinder. A Red Pitaya FPGA prototyping board running the MaRCoS framework was used to control the scanner components, and a GPA-FHDO amplifier board was used to drive the gradients. To simplify the scanner's operation, console software with an intuitive graphical user interface was developed in Python using the PyPulseq package for sequence calculations. Furthermore, the scanner was equipped with a cooling system, as well as options for passive and active shimming. After resolving several technical issues that arose during the assembly, the scanner is now able to acquire MR images with different sequences. While not suitable for real-world clinical applications, it can be utilized for educational purposes or as a low-cost prototyping platform. Moreover, it may serve as a reference design for future MRI development projects. All source code and resources are available on the project website mri4all.org, allowing other groups to replicate the scanner. EVIDENCE LEVEL: n/a TECHNICAL EFFICACY: Stage 1.

Low-dose oral minoxidil (LDOM) has emerged as a widely used off-label treatment for different types of alopecia, showing a favorable safety profile and effectiveness. Despite its growing use, it is essential to understand the possible associated adverse events (AEs) and their appropriate management to optimize this therapy. The aim of this article was to comprehensively review the AEs of LDOM treatment, describing their frequency, risk factors, affected anatomical sites, and management strategies. A search in the PubMed and EMBASE databases was performed for studies published before 31 December 2024, reporting the treatment of any type of hair loss with oral minoxidil. The most frequent AE is hypertrichosis, occurring in approximately 15% of patients, with a higher incidence in women and patients with higher doses. Fluid retention affects 1.3-10% of patients, particularly women, and typically occurs within 1-3 months of treatment. Other cardiovascular AEs, such as tachycardia or dizziness, occur in fewer than 5% of cases and are usually mild and transient. Severe AEs, including pericardial effusion, are extremely rare and often linked to compounding errors comprising an excessive dose. Management strategies include dose reduction, pharmacological interventions like diuretics for edema, and lifestyle measures such as sodium restriction. In most cases, AEs resolve without the need for treatment discontinuation. The favorable safety profile of LDOM makes it a valuable therapeutic option for alopecia, though careful patient selection, dose titration, and monitoring are essential to minimize risks.

The oral mucosal calcified nodule (OMCN) is a rare soft tissue lesion with only 7 cases reported in the English literature. It typically presents in the pediatric population as an asymptomatic submucosal nodule of less than 2 cm size affecting the maxillary ridge or palate, though other sites are reported. The histopathology displays stratified squamous epithelium overlying fibrous connective tissue with embedded calcified aggregates bordered by variable numbers of multinucleated giant cells. Surgical excision is curative. In this report, we present a new case of OMCN, outline the characteristic histopathologic features and review the cases reported in the English literature.

Progressive supranuclear palsy (PSP) is a rare, debilitating neurodegenerative disorder that significantly impairs both motor and cognitive functions. Current pharmacological treatments offer only transient symptomatic relief, driving interest in the past in alternative therapeutic strategies such as deep brain stimulation. Deep brain stimulation (DBS), known for its success in treating motor symptoms of Parkinson's disease, has been explored as a possible symptomatic treatment for PSP, considering the pedunculopontine nucleus (PPN), involved in motor control and postural stability, as a promising target for deep brain stimulation in PSP. However, its complex anatomy and the clinical variability of PSP complicate the prediction and generalization of the effectiveness of DBS. The present review examines the existing studies in the literature about DBS in PSP patients. Some studies highlighted modest benefits in motor symptoms, while others reported variable outcomes and inherent risks of the procedure. Generally, patients with a parkinsonism predominant phenotype have shown some subjective or clinical improvements in gait and balance when subjected to low-frequency stimulation. While DBS of the PPN holds promise for ameliorating gait and balance of PSP, current evidence does not yet establish clear criteria for ideal candidates, nor does it provide overwhelmingly supportive results in favor of PPN-DBS in PSP patients. Without any further systematic study is not possible to define accurate candidate selection parameters and understand long-term outcomes and safety profiles.

BACKGROUND AND AIMS/OBJECTIVE:The COVID-19 pandemic significantly decreased procedural volume in 2020, thus limiting training opportunities. First-year fellows (1YF) are particularly susceptible to reduction in endoscopic volume as they build foundational endoscopic skills. We used an objective validated tool, the Assessment of Competency in Endoscopy (ACE), to determine the effect of the pandemic on endoscopy competency in gastroenterology fellows. METHODS:This was a prospective cohort study conducted across two fellowship programs in New York City. Our primary outcome was the comparison of ACE scores of fellows starting gastroenterology fellowships in 2020 to the scores of those starting in 2016-2018. Our secondary outcome was to compare colonoscopy quality metrics and endoscopy volume between these cohorts. RESULTS:ACE scores were available from 11 pandemic and 10 pre-pandemic 1YFs, and five pandemic and 19 pre-pandemic second (2YF) and third-year fellows (3YF). ACE scores for 1YFs showed significantly higher rating in the pre-pandemic cohort for 4 domains including knowledge of indications and medical issues (p = 0.03), effective and efficient use of air, water, and suction (p = 0.04), pathology identification (p = 0.001), and overall hands-on skills (p = 0.004). 2YFs showed no significant differences. 3YFs showed significantly lower scores in the pandemic cohort in the effective and efficient use of air, water, and suction (p = 0.03), fellows' knowledge of therapeutic tools (p = 0.05), and fellows overall cognitive skills (p = 0.05). There were no significant differences in quality metrics between the cohorts, except longer procedure time for the pandemic cohort (p < 0.01). CONCLUSION/CONCLUSIONS:Following the COVID-19 pandemic, there was a decline in multiple aspects of endoscopic competency both at the beginning and the end of fellowship training. These trends are likely the result of a decrease in overall procedure volume early in training and highlight the need for supplementing trainee education with other modalities.

OBJECTIVE:Naming difficulty is a common symptom of left (i.e., language dominant) hemisphere epilepsy. As such, in the presurgical evaluation for drug-resistant epilepsy, which aims to localize the epileptogenic region, identification of a naming deficit typically implicates the left temporal region. However, the well-established finding of poor naming in those with left but not right (i.e., nondominant) hemisphere seizures in monolingual patients is unreliable in bilingual adults with epilepsy, despite proficiency in the language tested. We aimed to examine naming performance and its relation with seizure lateralization in bilingual children with epilepsy. METHODS:This multisite study included 57 bilingual and 202 monolingual pediatric epilepsy patients, aged 6-17 years. All patients underwent neuropsychological evaluation including assessment of auditory and visual object naming in English. RESULTS:In the context of age-appropriate English expressive vocabulary skills, bilingual children with epilepsy demonstrated significantly weaker auditory and visual naming than monolingual patients. Additionally, unlike monolingual patients, who showed poorer naming among those with left compared to those with right hemisphere seizures, bilingual children with unilateral left and right hemisphere seizures demonstrated similarly weak naming performances. Furthermore, naming score cutoffs failed to differentiate individual bilingual patients with left versus right hemisphere seizure onset as they did among monolingual patients. SIGNIFICANCE/CONCLUSIONS:Despite conversational proficiency and normal English expressive vocabulary, the relation between seizure laterality and naming performance demonstrated in monolingual children with unilateral seizures was not observed in a comparable group of bilingual children. Consequently, poor naming performance in bilingual children with epilepsy may be misinterpreted, most seriously in those with nondominant hemisphere seizures, as scores may be erroneously interpreted to reflect dominant hemisphere seizure involvement, potentially leading to unnecessary invasive and costly procedures. Results suggest cautious interpretation of naming performance in bilingual children with epilepsy.