Modeling electrical impedance in brain tissue with diffusion tensor imaging for functional neurosurgery applications.

Objective.Decades ago, neurosurgeons used electrical impedance measurements in the brain for coarse intraoperative tissue differentiation. Over time, these techniques were largely replaced by more refined imaging and electrophysiological localization. Today, advanced methods of diffusion tensor imaging (DTI) and finite element method (FEM) modeling may permit non-invasive, high-resolution intracerebral impedance prediction. However, expectations for tissue-impedance relationships and experimentally verified parameters for impedance modeling in human brains are lacking. This study seeks to address this need.Approach.We used FEM to simulate high-resolution single- and dual-electrode impedance measurements along linear electrode trajectories through (1) canonical gray and white matter tissue models, and (2) selected anatomic structures within whole-brain patient DTI-based models. We then compared intraoperative impedance measurements taken at known locations along deep brain stimulation (DBS) surgical trajectories with model predictions to evaluate model accuracy and refine model parameters.Main results.In DTI-FEM models, single- and dual-electrode configurations performed similarly. While only dual-electrode configurations were sensitive to white matter fiber orientation, other influences on impedance, such as white matter density, enabled single-electrode impedance measurements to display significant spatial variation even within purely white matter structures. We compared 308 intraoperative single-electrode impedance measurements in five DBS patients to DTI-FEM predictions at one-to-one corresponding locations. After calibration of model coefficients to these data, predicted impedances reliably estimated intraoperative measurements in all patients (R=0.784±0.116, n=5). Through this study, we derived an updated value for the slope coefficient of the DTI conductance model published by Tuch et al., k=0.0649 S·s/mm3(original k=0.844), for use specifically in humans at physiological frequencies.Significance.This is the first study to compare impedance estimates from imaging-based models of human brain tissue to experimental measurements at the same locations in vivo. Accurate, non-invasive, imaging-based impedance prediction has numerous applications in functional neurosurgery, including tissue mapping, intraoperative electrode localization, and DBS.

Optimize individualized energy delivery for septic patients using predictive deep learning models.

BACKGROUND AND OBJECTIVES: We aim to establish deep learning models to optimize the individualized energy delivery for septic patients. METHODS AND STUDY DESIGN: We conducted a study of adult septic patients in ICU, collecting 47 indicators for 14 days. We filtered out nutrition-related features and divided the data into datasets according to the three metabolic phases proposed by ESPEN: acute early, acute late, and rehabilitation. We then established optimal energy target models for each phase using deep learning and conducted external validation. RESULTS: A total of 179 patients in training dataset and 98 patients in external validation dataset were included in this study, and total data size was 3115 elements. The age, weight and BMI of the patients were 63.05 (95%CI 60.42-65.68), 61.31(95%CI 59.62-63.00) and 22.70 (95%CI 22.21-23.19), respectively. And 26.0% (72) of the patients were female. The models indicated that the optimal energy targets in the three phases were 900kcal/d, 2300kcal/d, and 2000kcal/d, respectively. Excessive energy intake increased mortality rapidly in the early period of the acute phase. Insufficient energy in the late period of the acute phase significantly raised the mortality as well. For the rehabilitation phase, too much or too little energy delivery were both associated with elevated death risk. CONCLUSIONS: Our study established time-series prediction models for septic patients to optimize energy delivery in the ICU. We recommended permissive underfeeding only in the early acute phase. Later, increased energy intake may improve survival and settle energy debts caused by underfeeding.

Initial Management of BRAF V600E-Variant Anaplastic Thyroid Cancer: The FAST Multidisciplinary Group Consensus Statement.

Importance: BRAF/MEK inhibitors revolutionized the treatment of BRAF V600E-variant anaplastic thyroid carcinoma (BRAFv-ATC), offering improved outcomes for patients with this previously incurable disease. Observations: Anaplastic thyroid carcinoma (ATC) accounts for approximately half of thyroid cancer-related deaths. It presents as a rapidly growing tumor that often invades locoregional structures and spreads to distant sites early; therefore, prompt diagnosis, staging, and treatment initiation are of the essence in the treatment of ATC. Although most oncologists will encounter a patient with ATC in their practice, the rarity of this disease makes treatment challenging, particularly because those with BRAFv-ATC no longer have a dismal prognosis. BRAF/MEK kinase inhibitors have transformed the outlook and treatment of BRAFv-ATC. Therefore, molecular profiling to identify these patients is critical. More recently, the addition of immunotherapy to BRAF/MEK inhibitors as well as the use of the neoadjuvant approach were shown to further improve survival outcomes in BRAFv-ATC. Many of these recent advances have not yet been incorporated in the currently available guidelines, allowing for disparities in the treatment of patients with BRAFv-ATC across the US. With the increasing complexity in the management of BRAFv-ATC, this Consensus Statement aims to formulate guiding recommendations from a group of experts to facilitate therapeutic decision-making. Conclusions and Relevance: This Consensus Statement from the FAST (Facilitating Anaplastic Thyroid Cancer Specialized Treatment) group at MD Anderson Cancer Center emphasizes that rapid identification of a BRAF V600E pathogenic variant and timely initiation of sequential therapy are critical to avoid excess morbidity and mortality in patients with BRAFv-ATC. In the past decade, remarkable progress has been made in the treatment of patients with BRAFv-ATC, justifying these new evidence-based recommendations reached through a consensus of experts from a high-volume center.

Atropine as an Adjunct in the Management of Pediatric Patients With Hypertrophic Pyloric Stenosis: A Single Institution Experience and Review of the Literature.

Infantile hypertrophic pyloric stenosis (IHPS) is a condition whereby there is a thickening of the pyloric muscle, leading to obstruction of the gastric outflow. Typically present within three to five weeks of life, it presents as postprandial non-bilious projectile vomiting. Commonly, a pyloromyotomy is the gold standard to relieve the obstruction. However, in a subset of patients not amenable to undergo surgery or anesthesia, or for postoperative persistent or recurrent obstruction, atropine may offer an alternative treatment. A retrospective review was performed on pediatric patients with hypertrophic pyloric stenosis utilizing the electronic medical record. Data included were demographics, workup data, treatment, outcomes, and symptom resolution. Approval was obtained by the institutional review board of the host institution. Five pediatric patients, with an average age of 2.1 months, received atropine treatment for IHPS. The average time to reach full feeds since the initiation of atropine was approximately four days. Three of the five patients were successfully managed with IV atropine, which was then transitioned to oral atropine and tapered off as outpatients, leading to the resolution of symptoms. The remaining two patients were considered failures of medical management and subsequently required surgery. Atropine use as an alternative treatment for IHPS may be considered when patients are not able to undergo surgery or anesthesia or have recurrent or persistent obstructive symptoms postoperatively. In this limited study, atropine was found to be safe and effective. Randomized controlled studies may lend additional merit to this therapy in the future.

Privacy preservation for federated learning in health care.

Artificial intelligence (AI) shows potential to improve health care by leveraging data to build models that can inform clinical workflows. However, access to large quantities of diverse data is needed to develop robust generalizable models. Data sharing across institutions is not always feasible due to legal, security, and privacy concerns. Federated learning (FL) allows for multi-institutional training of AI models, obviating data sharing, albeit with different security and privacy concerns. Specifically, insights exchanged during FL can leak information about institutional data. In addition, FL can introduce issues when there is limited trust among the entities performing the compute. With the growing adoption of FL in health care, it is imperative to elucidate the potential risks. We thus summarize privacy-preserving FL literature in this work with special regard to health care. We draw attention to threats and review mitigation approaches. We anticipate this review to become a health-care researcher's guide to security and privacy in FL.

Building a translational cancer dependency map for The Cancer Genome Atlas.

Cancer dependency maps have accelerated the discovery of tumor vulnerabilities that can be exploited as drug targets when translatable to patients. The Cancer Genome Atlas (TCGA) is a compendium of 'maps' detailing the genetic, epigenetic and molecular changes that occur during the pathogenesis of cancer, yet it lacks a dependency map to translate gene essentiality in patient tumors. Here, we used machine learning to build translational dependency maps for patient tumors, which identified tumor vulnerabilities that predict drug responses and disease outcomes. A similar approach was used to map gene tolerability in healthy tissues to prioritize tumor vulnerabilities with the best therapeutic windows. A subset of patient-translatable synthetic lethalities were experimentally tested, including PAPSS1/PAPSS12 and CNOT7/CNOT78, which were validated in vitro and in vivo. Notably, PAPSS1 synthetic lethality was driven by collateral deletion of PAPSS2 with PTEN and was correlated with patient survival. Finally, the translational dependency map is provided as a web-based application for exploring tumor vulnerabilities.

Mutational Characteristics and Clinical Outcomes for Lung Adenocarcinoma With EGFR Germline Mutations.

INTRODUCTION: Germline mutations driving lung cancer have been infrequently reported in the literature, with EGFR T790M being a known germline mutation identified in 1% of NSCLCs. Typically, a somatic EGFR mutation is acquired to develop lung adenocarcinoma. Osimertinib has become a standard-of-care treatment for EGFR T790M-positive lung cancer. METHODS: We perform a retrospective analysis through the Lung Cancer Moon Shot GEMINI database at the University of Texas MD Anderson Cancer Center. Of the patients that underwent cell-free DNA analysis, germline mutations were identified by those with high variant allelic fraction approximating 50%, followed by further confirmation on genetic testing. RESULTS: We identified 22 patients with germline EGFR mutations, with the majority harboring an EGFR T790M mutation (95.5%) and an EGFR L858R somatic mutation (50%). Notably, most patients were female (86.4%), non-smokers (81.8%), white (86.4%), had a family history of lung cancer (59.1%), and stage IV at diagnosis (72.7%). A distinct radiographic pattern of small multifocal ground-glass pulmonary nodules was observed in the majority of our cohort (72.7%). Among the 18 with advanced-stage NSCLC, 12 patients (66.7%) were treated with first-line osimertinib, demonstrating a median progression-free survival (PFS) of 16.9 months (95% confidence interval [CI]: 6.3-not reached [NR]). Others were treated with first-line afatinib (11.1%) or chemotherapy (22.2%). Among the 17 patients treated with osimertinib (in first or second-line), median PFS was 20.4 months (95% CI: 6.3-NR) and median overall survival was 82.0 months (95% CI: 28.4-NR). CONCLUSIONS: Based on our institutional cohort, NSCLC driven by EGFR germline mutations occurs more frequently in non-smoking, white females with multi-focal pulmonary nodules radiographically. Osimertinib for advanced germline EGFR-mutated NSCLC renders similar PFS compared to somatic T790M EGFR-mutated NSCLC.

Assessing readiness to use electronic health record data for outcome ascertainment in clinical trials - A case study.

BACKGROUND: Variable data quality poses a challenge to using electronic health record (EHR) data to ascertain acute clinical outcomes in multi-site clinical trials. Differing EHR platforms and data comprehensiveness across clinical trial sites, especially if patients received care outside of the clinical site's network, can also affect validity of results. Overcoming these challenges requires a structured approach. METHODS: We propose a framework and create a checklist to assess the readiness of clinical sites to contribute EHR data to a clinical trial for the purpose of outcome ascertainment, based on our experience with the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) study, which enrolled 5451 participants in 86 primary care practices across 10 healthcare systems (sites). RESULTS: The site readiness checklist includes assessment of the infrastructure (i.e., size and structure of the site's healthcare system or clinical network), data procurement (i.e., quality of the data), and cost of obtaining study data. The checklist emphasizes the importance of understanding how data are captured and integrated across a site's catchment area and having a protocol in place for data procurement to ensure consistent and uniform extraction across each site. CONCLUSIONS: We suggest rigorous, prospective vetting of the data quality and infrastructure of each clinical site before launching a multi-site trial dependent on EHR data. The proposed checklist serves as a guiding tool to help investigators ensure robust and unbiased data capture for their clinical trials. ORIGINAL TRIAL REGISTRATION NUMBER: NCT02475850.

Efficacy of taurine-enhanced enteral nutrition in improving the outcomes of critically ill patients: A systematic review and meta-analysis.

BACKGROUND: Taurine is considered an immunomodulatory agent. From current reports on clinical studies, we conducted a systematic review and meta-analysis to investigate the effects of taurine-enhanced enteral nutrition (EN) on the outcomes of critically ill patients to resolve conflicting evidence in literature. METHODS: Literature from PubMed, EMBASE, Web of Science, Cochrane Library, CNKI, SINOMED, and WanFang databases were retrieved, and randomized controlled trials (RCTs) were identified. The time range spanned from January 1, 2000, to January 31, 2024. The Cochrane Collaboration Tool was used to evaluate the risk of bias. We used the GRADE approach to rate the quality of evidence and the I2 test to assess the statistical heterogeneity of the results. Risk ratio (RR), mean difference (MD), and 95% confidence interval (95% CI) were used to analyze measurement data. RESULTS: Four trials involving 236 patients were finally included. The meta-analysis results indicated that taurine-enhanced EN did not reduce mortality (RR = 0.70, p = 0.45, 95% CI [0.28, 1.80], two trials, 176 participants, low quality). There was also no significant difference in length of stay in the intensive care unit (ICU) between the taurine-enhanced EN and control groups. Taurine-enhanced EN may reduce pro-inflammatory factor interleukin-6 (IL-6) levels in critically ill patients（the result about IL-6 cannot be pooled）. However, taurine-enhanced EN had no significant impact on high-sensitivity-C-reactive protein levels (MD = -0.41, p = 0.40, 95% CI [-1.35, 0.54], two trials, 60 participants, low quality). DISCUSSION: Taurine-enhanced EN may reduce IL-6 levels and is not associated with improved clinical outcomes in critically ill patients, which may have potential immunoregulatory effects in critically ill patients. Given that published studies have small samples, the above conclusions need to be verified by more rigorously designed large-sample clinical trials.

Artificial intelligence driven malnutrition diagnostic model for patients with acute abdomen based on GLIM criteria: a cross-sectional research protocol.

BACKGROUND: Patients with acute abdomen often experience reduced voluntary intake and a hypermetabolic process, leading to a high occurrence of malnutrition. The Global Leadership Initiative on Malnutrition (GLIM) criteria have rapidly developed into a principal methodological tool for nutritional diagnosis. Additionally, machine learning is emerging to establish artificial intelligent-enabled diagnostic models, but the accuracy and robustness need to be verified. We aimed to establish an intelligence-enabled malnutrition diagnosis model based on GLIM for patients with acute abdomen. METHOD: This study is a single-centre, cross-sectional observational investigation into the prevalence of malnutrition in patients with acute abdomen using the GLIM criteria. Data collection occurs on the day of admission, at 3 and 7 days post-admission, including biochemical analysis, body composition indicators, disease severity scoring, nutritional risk screening, malnutrition diagnosis and nutritional support information. The occurrence rate of malnutrition in patients with acute abdomen is analysed with the GLIM criteria based on the Nutritional Risk Screening 2002 and the Mini Nutritional Assessment Short-Form to investigate the sensitivity and accuracy of the GLIM criteria. After data cleansing and preprocessing, a machine learning approach is employed to establish a predictive model for malnutrition diagnosis in patients with acute abdomen based on the GLIM criteria. ETHICS AND DISSEMINATION: This study has obtained ethical approval from the Ethics Committee of the Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital on 28 November 2022 (Yan-2022-442). The results of this study will be disseminated in peer-reviewed journals, at scientific conferences and directly to study participants. TRIAL REGISTRATION NUMBER: ChiCTR2200067044.

Survival outcomes and toxicity of adjuvant immunotherapy after definitive concurrent chemotherapy with proton beam radiation therapy for patients with inoperable locally advanced non-small cell lung carcinoma.

INTRODUCTION: Adjuvant immunotherapy (IO) following concurrent chemotherapy and photon radiation therapy confers an overall survival (OS) benefit for patients with inoperable locally advanced non-small cell lung carcinoma (LA-NSCLC); however, outcomes of adjuvant IO after concurrent chemotherapy with proton beam therapy (CPBT) are unknown. We investigated OS and toxicity after CPBT with adjuvant IO versus CPBT alone for inoperable LA-NSCLC. MATERIALS AND METHODS: We analyzed 354 patients with LA-NSCLC who were prospectively treated with CPBT with or without adjuvant IO from 2009 to 2021. Optimal variable ratio propensity score matching (PSM) matched CPBT with CPBT + IO patients. Survival was estimated with the Kaplan-Meier method and compared with log-rank tests. Multivariable Cox proportional hazards regression evaluated the effect of IO on disease outcomes. RESULTS: Median age was 70 years; 71 (20%) received CPBT + IO and 283 (80%) received CPBT only. After PSM, 71 CPBT patients were matched with 71 CPBT + IO patients. Three-year survival rates for CPBT + IO vs CPBT were: OS 67% vs 30% (P < 0.001) and PFS 59% vs 35% (P = 0.017). Three-year LRFS (P = 0.137) and DMFS (P = 0.086) did not differ. Receipt of adjuvant IO was a strong predictor of OS (HR 0.40, P = 0.001) and PFS (HR 0.56, P = 0.030), but not LRFS (HR 0.61, P = 0.121) or DMFS (HR 0.61, P = 0.136). There was an increased incidence of grade ≥3 esophagitis in the CPBT-only group (6% CPBT + IO vs 17% CPBT, P = 0.037). CONCLUSION: This study, one of the first to investigate CPBT followed by IO for inoperable LA-NSCLC, showed that IO conferred survival benefits with no increased rates of toxicity.

The chromatin landscape of healthy and injured cell types in the human kidney.

There is a need to define regions of gene activation or repression that control human kidney cells in states of health, injury, and repair to understand the molecular pathogenesis of kidney disease and design therapeutic strategies. Comprehensive integration of gene expression with epigenetic features that define regulatory elements remains a significant challenge. We measure dual single nucleus RNA expression and chromatin accessibility, DNA methylation, and H3K27ac, H3K4me1, H3K4me3, and H3K27me3 histone modifications to decipher the chromatin landscape and gene regulation of the kidney in reference and adaptive injury states. We establish a spatially-anchored epigenomic atlas to define the kidney's active, silent, and regulatory accessible chromatin regions across the genome. Using this atlas, we note distinct control of adaptive injury in different epithelial cell types. A proximal tubule cell transcription factor network of ELF3, KLF6, and KLF10 regulates the transition between health and injury, while in thick ascending limb cells this transition is regulated by NR2F1. Further, combined perturbation of ELF3, KLF6, and KLF10 distinguishes two adaptive proximal tubular cell subtypes, one of which manifested a repair trajectory after knockout. This atlas will serve as a foundation to facilitate targeted cell-specific therapeutics by reprogramming gene regulatory networks.

Genome-scale functional genomics screening highlights genes impacting protein fucosylation in Chinese hamster ovary cells.

N-linked glycosylation is a common post-translational modification that has various effects on multiple types of proteins. The extent to which an N-linked glycoprotein is modified and the identity of glycans species involved is of great interest to the biopharmaceutical industry, since glycosylation can impact the efficacy and safety of therapeutic monoclonal antibodies (mAbs). mAbs lacking core fucose, for example, display enhanced clinical efficacy through increased antibody-dependent cellular cytotoxicity. We performed a genome-wide CRISPR knockout screen in Chinese hamster ovary (CHO) cells, the workhorse cell culture system for industrial production of mAbs, aimed at identifying novel regulators of protein fucosylation. Using a lectin binding assay, we identified 224 gene perturbations that significantly alter protein fucosylation, including well-known glycosylation genes. This functional genomics framework could readily be extended and applied to study the genetic pathways involved in regulation of other glycoforms. We hope this resource will provide useful guidance toward the development of next generation CHO cell lines and mAb therapeutics.

Electrical Stimulation of Regenerative Peripheral Nerve Interfaces (RPNIs) Induces Referred Sensations in People With Upper Limb Loss.

Individuals with upper limb loss lack sensation of the missing hand, which can negatively impact their daily function. Several groups have attempted to restore this sensation through electrical stimulation of residual nerves. The purpose of this study was to explore the utility of regenerative peripheral nerve interfaces (RPNIs) in eliciting referred sensation. In four participants with upper limb loss, we characterized the quality and location of sensation elicited through electrical stimulation of RPNIs over time. We also measured functional stimulation ranges (sensory perception and discomfort thresholds), sensitivity to changes in stimulation amplitude, and ability to differentiate objects of different stiffness and sizes. Over a period of up to 54 months, stimulation of RPNIs elicited sensations that were consistent in quality (e.g. tingling, kinesthesia) and were perceived in the missing hand and forearm. The location of elicited sensation was partially-stable to stable in 13 of 14 RPNIs. For 5 of 7 RPNIs tested, participants demonstrated a sensitivity to changes in stimulation amplitude, with an average just noticeable difference of 45 nC. In a case study, one participant was provided RPNI stimulation proportional to prosthetic grip force. She identified four objects of different sizes and stiffness with 56% accuracy with stimulation alone and 100% accuracy when stimulation was combined with visual feedback of hand position. Collectively, these experiments suggest that RPNIs have the potential to be used in future bi-directional prosthetic systems.

Dorsal subthalamic deep brain stimulation improves pain in Parkinson's disease.

Introduction: Inconsistent effects of subthalamic deep brain stimulation (STN DBS) on pain, a common non-motor symptom of Parkinson's disease (PD), may be due to variations in active contact location relative to some pain-reducing locus of stimulation. This study models and compares the loci of maximal effect for pain reduction and motor improvement in STN DBS. Methods: We measured Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS) Part I pain score (item-9), and MDS-UPDRS Part III motor score, preoperatively and 6-12 months after STN DBS. An ordinary least-squares regression model was used to examine active contact location as a predictor of follow-up pain score while controlling for baseline pain, age, dopaminergic medication, and motor improvement. An atlas-independent isotropic electric field model was applied to distinguish sites of maximally effective stimulation for pain and motor improvement. Results: In 74 PD patients, mean pain score significantly improved after STN DBS (p = 0.01). In a regression model, more dorsal active contact location was the only significant predictor of pain improvement (R2 = 0.17, p = 0.03). The stimulation locus for maximal pain improvement was lateral, anterior, and dorsal to that for maximal motor improvement. Conclusion: STN stimulation, dorsal to the site of optimal motor improvement, improves pain. This region contains the zona incerta, which is known to modulate pain in humans, and may explain this observation.

The chromatin landscape of healthy and injured cell types in the human kidney.

There is a need to define regions of gene activation or repression that control human kidney cells in states of health, injury, and repair to understand the molecular pathogenesis of kidney disease and design therapeutic strategies. However, comprehensive integration of gene expression with epigenetic features that define regulatory elements remains a significant challenge. We measured dual single nucleus RNA expression and chromatin accessibility, DNA methylation, and H3K27ac, H3K4me1, H3K4me3, and H3K27me3 histone modifications to decipher the chromatin landscape and gene regulation of the kidney in reference and adaptive injury states. We established a comprehensive and spatially-anchored epigenomic atlas to define the kidney's active, silent, and regulatory accessible chromatin regions across the genome. Using this atlas, we noted distinct control of adaptive injury in different epithelial cell types. A proximal tubule cell transcription factor network of ELF3 , KLF6 , and KLF10 regulated the transition between health and injury, while in thick ascending limb cells this transition was regulated by NR2F1 . Further, combined perturbation of ELF3 , KLF6 , and KLF10 distinguished two adaptive proximal tubular cell subtypes, one of which manifested a repair trajectory after knockout. This atlas will serve as a foundation to facilitate targeted cell-specific therapeutics by reprogramming gene regulatory networks.

Granulomatous mastitis secondary to Corynebacterium requiring surgical intervention: a complicated diagnosis.

Corynebacterium species is a Gram-positive bacillus endogenous to human integument that has previously been associated with idiopathic granulomatous mastitis. The diagnosis and treatment of this bacteria may be complicated by inability to distinguish colonization from contamination and infection. We present an uncommon case of granulomatous mastitis associated with negative wound cultures requiring surgical intervention.

Collaborative Privacy-preserving Approaches for Distributed Deep Learning Using Multi-Institutional Data.

Deep learning (DL) algorithms have shown remarkable potential in automating various tasks in medical imaging and radiologic reporting. However, models trained on low quantities of data or only using data from a single institution often are not generalizable to other institutions, which may have different patient demographics or data acquisition characteristics. Therefore, training DL algorithms using data from multiple institutions is crucial to improving the robustness and generalizability of clinically useful DL models. In the context of medical data, simply pooling data from each institution to a central location to train a model poses several issues such as increased risk to patient privacy, increased costs for data storage and transfer, and regulatory challenges. These challenges of centrally hosting data have motivated the development of distributed machine learning techniques and frameworks for collaborative learning that facilitate the training of DL models without the need to explicitly share private medical data. The authors describe several popular methods for collaborative training and review the main considerations for deploying these models. They also highlight publicly available software frameworks for federated learning and showcase several real-world examples of collaborative learning. The authors conclude by discussing some key challenges and future research directions for distributed DL. They aim to introduce clinicians to the benefits, limitations, and risks of using distributed DL for the development of medical artificial intelligence algorithms. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.