Comparative analysis of glymphatic system alterations in multiple sclerosis and neuromyelitis optica spectrum disorder using MRI indices from diffusion tensor imaging.

OBJECTIVE: The glymphatic system is a glial-based perivascular network that promotes brain metabolic waste clearance. Glymphatic system dysfunction has been observed in both multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD), indicating the role of neuroinflammation in the glymphatic system. However, little is known about how the two diseases differently affect the human glymphatic system. The present study aims to evaluate the diffusion MRI-based measures of the glymphatic system by contrasting MS and NMOSD. METHODS: This prospective study included 63 patients with NMOSD (n = 21) and MS (n = 42) who underwent DTI. The fractional volume of extracellular-free water (FW) and an index of diffusion tensor imaging (DTI) along the perivascular space (DTI-ALPS) were used as indirect indicators of water diffusivity in the interstitial extracellular and perivenous spaces of white matter, respectively. Age and EDSS scores were adjusted. RESULTS: Using Bayesian hypothesis testing, we show that the present data substantially favor the null model of no differences between MS and NMOSD for the diffusion MRI-based measures of the glymphatic system. The inclusion Bayes factor (BF10) of model-averaged probabilities of the group (MS, NMOSD) was 0.280 for FW and 0.236 for the ALPS index. CONCLUSION: Together, these findings suggest that glymphatic alteration associated with MS and NMOSD might be similar and common as an eventual result, albeit the disease etiologies differ. PRACTITIONER POINTS: Previous literature indicates important glymphatic system alteration in MS and NMOSD. We explore the difference between MS and NMOSD using diffusion MRI-based measures of the glymphatic system. We show support for the null hypothesis of no difference between MS and NMOSD. This suggests that glymphatic alteration associated with MS and NMOSD might be similar and common etiology.

Prevalence, incidence, and treatment trends of adenomyosis in South Korean women for 15 years: A national population-based study.

OBJECTIVE: Adenomyosis is associated with female infertility worldwide. With improvements in imaging methods, such as pelvic magnetic resonance imaging, the diagnosis and treatment of adenomyosis have changed. This study aimed to evaluate the overall prevalence, incidence, and treatment trends of adenomyosis in South Korea using data from the Korean National Health Insurance Service Database (NHIS). METHODS: Data were collected from the Korean NHIS, a population-based complete enumeration database. A total of 678 641 women aged 11-55 years diagnosed with adenomyosis (N80.0 ICD-10 code) from the database from 2002 to 2016 were enrolled. After applying a one-year look-back method, 629 592 patients were analyzed to estimate the prevalence, incidence, and treatment trends of adenomyosis. RESULTS: The overall prevalence during the study period was 3.86 per 1000 people. The prevalence of adenomyosis has increased from 1.42 per 1000 individuals in 2002 to 7.50 per 1000 individuals in 2016. The crude annual incidence rate of adenomyosis was 1.62 per 1000 people in 2003, which increased to 4.12 per 1000 people in 2016. In addition, the proportion of uterus-preserving surgeries in adenomyosis treatments has increased from 7.51% to 21.29% over 15 years. CONCLUSION: The prevalence and incidence of adenomyosis in South Korea increased between 2002 and 2016. Furthermore, the proportion of uterus-preserving surgeries and progestin prescriptions for adenomyosis treatment has increased. We expect that our findings will raise awareness of the necessity for fertility preservation through earlier diagnosis and proper management of patients with adenomyosis.

Harnessing the Power of IL-7 to Boost T Cell Immunity in Experimental and Clinical Immunotherapies.

The cytokine IL-7 plays critical and nonredundant roles in T cell immunity so that the abundance and availability of IL-7 act as key regulatory mechanisms in T cell immunity. Importantly, IL-7 is not produced by T cells themselves but primarily by non-lymphoid lineage stromal cells and epithelial cells that are limited in their numbers. Thus, T cells depend on cell extrinsic IL-7, and the amount of in vivo IL-7 is considered a major factor in maximizing and maintaining the number of T cells in peripheral tissues. Moreover, IL-7 provides metabolic cues and promotes the survival of both naïve and memory T cells. Thus, IL-7 is also essential for the functional fitness of T cells. In this regard, there has been an extensive effort trying to increase the protein abundance of IL-7 in vivo, with the aim to augment T cell immunity and harness T cell functions in anti-tumor responses. Such approaches started under experimental animal models, but they recently culminated into clinical studies, with striking effects in re-establishing T cell immunity in immunocompromised patients, as well as boosting anti-tumor effects. Depending on the design, glycosylation, and the structure of recombinantly engineered IL-7 proteins and their mimetics, recombinant IL-7 molecules have shown dramatic differences in their stability, efficacy, cellular effects, and overall immune functions. The current review is aimed to summarize the past and present efforts in the field that led to clinical trials, and to highlight the therapeutical significance of IL-7 biology as a master regulator of T cell immunity.

Improved Oral Health Is Associated with a Lower Risk of Late Onset Ankylosing Spondylitis: A Nationwide Cohort Study.

Background: The aim of this study was to evaluate the association of oral health status and habits with the occurrence of ankylosing spondylitis (AS) in a nationwide population-based cohort in a longitudinal setting. Methods: A total of 2,415,963 individuals aged 40-79 years who underwent oral health examinations were included from the National Health Insurance Service-National Health Screening (NHIS-HEALS) cohort of Korea between 2003 and 2004. The occurrence of AS was analyzed according to the oral health status and oral hygiene habits. Results: Among 2,271,221 of the participants, AS occurred in 6366 (0.3%) participants over 16.7 years. The likelihood of AS was higher in participants who had periodontitis (hazard ratio [HR]: 1.33, 95% confidence interval [CI]: 1.20-1.46, p < 0.0001) and more missing teeth (HR: 1.68, 95% CI: 1.42-1.99, p < 0.0001). However, better oral hygiene habits such as frequent tooth brushing (HR: 0.77, 95% CI: 0.71-0.83, p < 0.0001) and a history of dental scaling within the last year (HR 0.88, 95% CI 0.82-0.95, p = 0.001) were associated with a lower occurrence of AS. Conclusions: Periodontitis and an increased number of missing teeth could be related to the occurrence of late-onset AS. Improved oral hygiene care may attenuate the likelihood of late-onset AS.

Cutting-Edge HEK293T Protein-Integrated Lipid Nanostructures: Boosting Biocompatibility and Efficacy.

Recently, artificial exosomes have been developed to overcome the challenges of natural exosomes, such as production scalability and stability. In the production of artificial exosomes, the incorporation of membrane proteins into lipid nanostructures is emerging as a notable approach for enhancing biocompatibility and treatment efficacy. This study focuses on incorporating HEK293T cell-derived membrane proteins into liposomes to create membrane-protein-bound liposomes (MPLCs), with the goal of improving their effectiveness as anticancer therapeutics. MPLCs were generated by combining two key elements: lipid components that are identical to those in conventional liposomes (CLs) and membrane protein components uniquely derived from HEK293T cells. An extensive comparison of CLs and MPLCs was conducted across multiple in vitro and in vivo cancer models, employing advanced techniques such as cryo-TEM (tramsmission electron microscopy) imaging and FT-IR (fourier transform infrared spectroscopy). MPLCs displayed superior membrane fusion capabilities in cancer cell lines, with significantly higher cellular uptake. Additionally, MPLCs maintained their morphology and size better than CLs when exposed to FBS (fetal bovine serum), suggesting enhanced serum stability. In a xenograft mouse model using HeLa and ASPC cancer cells, intravenous administration of MPLCs MPLCs accumulated more in tumor tissues, highlighting their potential for targeted cancer therapy. Overall, these results indicate that MPLCs have superior tumor-targeting properties, possibly attributable to their membrane protein composition, offering promising prospects for enhancing drug delivery efficiency in cancer treatments. This research could offer new clinical application opportunities, as it uses MPLCs with membrane proteins from HEK293T cells, which are known for their efficient production and compatibility with GMP (good manufacturing practice) standards.

Generation of an induced pluripotent stem cell line (KNIHi001-A) by reprogramming peripheral blood mononuclear cells isolated from a patient with Parkinson's disease.

Parkinson's disease is a degenerative brain disorder characterized by dopamine neuronal degeneration and dopamine transporter loss. In this study, we generated an induced pluripotent stem cell (iPSC) line, KNIHi001-A, from the peripheral blood mononuclear cells (PBMCs) of a 76-year-old man with Parkinson's disease. The non-integrating Sendai virus was used to reprogram iPSCs. iPSCs exhibit pluripotent markers, a normal karyotype, viral clearance, and the ability to differentiate into the three germ layers.

Pathways and mechanisms of CD4+CD8αα+ intraepithelial T cell development.

The mammalian small intestine epithelium harbors a peculiar population of CD4+CD8αα+ T cells that are derived from mature CD4+ T cells through reprogramming of lineage-specific transcription factors. CD4+CD8αα+ T cells occupy a unique niche in T cell biology because they exhibit mixed phenotypes and functional characteristics of both CD4+ helper and CD8+ cytotoxic T cells. The molecular pathways driving their generation are not fully mapped. However, recent studies demonstrate the unique role of the commensal gut microbiota as well as distinct cytokine and chemokine requirements in the differentiation and survival of these cells. We review the established and newly identified factors involved in the generation of CD4+CD8αα+ intraepithelial lymphocytes (IELs) and place them in the context of the molecular machinery that drives their phenotypic and functional differentiation.

Pioneering PGC-1α-boosted secretome: a novel approach to combating liver fibrosis.

Purpose: Liver fibrosis is a critical health issue with limited treatment options. This study investigates the potential of PGC-Sec, a secretome derived from peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α)-overexpressing adipose-derived stem cells (ASCs), as a novel therapeutic strategy for liver fibrosis. Methods: Upon achieving a cellular confluence of 70%-80%, ASCs were transfected with pcDNA-PGC-1α. PGC-Sec, obtained through concentration of conditioned media using ultrafiltration units with a 3-kDa cutoff, was assessed through in vitro assays and in vitro mouse models. Results: In vitro, PGC-Sec significantly reduced LX2 human hepatic stellate cell proliferation and mitigated mitochondrial oxidative stress compared to the control-secretome. In an in vivo mouse model, PGC-Sec treatment led to notable reductions in hepatic enzyme activity, serum proinflammatory cytokine concentrations, and fibrosis-related marker expression. Histological analysis demonstrated improved liver histology and reduced fibrosis severity in PGC-Sec-treated mice. Immunohistochemical staining confirmed enhanced expression of PGC-1α, optic atrophy 1 (a mitochondrial function marker), and peroxisome proliferator-activated receptor alpha (an antifibrogenic marker) in the PGC-Sec-treated group, along with reduced collagen type 1A expression (a profibrogenic marker). Conclusion: These findings highlight the therapeutic potential of PGC-Sec in combating liver fibrosis by enhancing mitochondrial biogenesis and function, and promoting antifibrotic processes. PGC-Sec holds promise as a novel treatment strategy for liver fibrosis.

Effect of information provision on parental intention toward COVID-19 vaccination for children: a nationwide survey experiment.

The reluctance of parents to vaccinate their children against COVID-19 was prevalent particularly when uncertainty over vaccination outcomes prevailed. We conducted a nationwide randomized online survey experiment to assess the effect of information provision on parental intention for COVID-19 vaccination before the government started vaccination for children in South Korea. Parents of elementary school children were provided with either no information (Control), information on vaccine profile (vaccine informed group; VI), or COVID-19 (disease informed group; DI). Among 359,110 participants, parental intention for vaccination of children was significantly higher in both VI and DI groups compared with the Control group. In terms of effect size, information on COVID-19 vaccine increased likelihood to vaccinate by 1620 per 100,000 parents and reduced vaccine hesitancy by 1340 per 100,000 parents. Consistently with the positive effect on vaccination intention, both VI and DI interventions increased participants' perceptions on vaccination benefits being higher than its risks and vaccination risks being lower than health risks of COVID-19 infection, and self-reported trust in COVID-19 information. Our results lend strong support to the claim that the provision of targeted, tailored information on COVID-19 vaccine and infection increases parental intention to vaccinate children and reduces vaccine hesitancy.

The ever-expanding role of cytokine receptor DR3 in T cells.

Death Receptor 3 (DR3) is a cytokine receptor of the Tumor Necrosis Factor receptor superfamily that plays a multifaceted role in both innate and adaptive immunity. Based on the death domain motif in its cytosolic tail, DR3 had been proposed and functionally affirmed as a trigger of apoptosis. Further studies, however, also revealed roles of DR3 in other cellular pathways, including inflammation, survival, and proliferation. DR3 is expressed in various cell types, including T cells, B cells, innate lymphocytes, myeloid cells, fibroblasts, and even outside the immune system. Because DR3 is mainly expressed on T cells, DR3-mediated immune perturbations leading to autoimmunity and other diseases were mostly attributed to DR3 activation of T cells. However, which T cell subset and what T effector functions are controlled by DR3 to drive these processes remain incompletely understood. DR3 engagement was previously found to alter CD4 T helper subset differentiation, expand the Foxp3+ Treg cell pool, and maintain intraepithelial γδ T cells in the gut. Recent studies further unveiled a previously unacknowledged aspect of DR3 in regulating innate-like invariant NKT (iNKT) cell activation, expanding the scope of DR3-mediated immunity in T lineage cells. Importantly, in the context of iNKT cells, DR3 ligation exerted costimulatory effects in agonistic TCR signaling, unveiling a new regulatory framework in T cell activation and proliferation. The current review is aimed at summarizing such recent findings on the role of DR3 on conventional T cells and innate-like T cells and discussing them in the context of immunopathogenesis.

Effectiveness of live-streaming tele-exercise intervention in patients with Parkinson's disease: A pilot study.

Introduction: Exercise can improve both motor and non-motor symptoms in people with Parkinson's disease (PwP), but there is an unmet need of accessible and sustainable exercise options. This study aimed to evaluate the effect, feasibility, and safety of a regularly performed live-streaming tele-exercise intervention for PwP. Methods: A live-streaming exercise intervention was implemented twice a week for 12 weeks in PwP. We measured the motor and nonmotor scales in these patients before and after the intervention. Changes in clinical scores from baseline to post-intervention were analyzed using a paired t-test. Factors associated with improvements in clinical scales and compliance were analyzed using Pearson's correlation analysis. Results: 56 participants were enrolled in the study. There were significant improvements in HADS-A (p = 0.007), HADS-D (p < 0.001), UPDRS part III (p < 0.001), UPDRS total (p = 0.015), H&Y stage (p = 0.027), and PFS-16 (p = 0.026) scores following intervention. Motor improvements were associated with improvements in mood symptoms and fatigue. Higher motor impairment at baseline was associated with a higher compliance rate and better composite motor and non-motor outcomes (ΔUPDRS total score) post-intervention. Overall, the 12-week tele-exercise program was feasible and safe for PwP. No adverse event was reported. Overall adherence was 60.0% in our cohort, and 83.4% were able to participate in more than half of the exercise routines. Conclusion: The live-streaming tele-exercise intervention is a safe, feasible, and effective non-pharmacological treatment option that can alleviate fatigue and improve mood and motor symptoms in PwP.

Sodium Intake Estimation in Hospital Patients Using AI-Based Imaging: Prospective Pilot Study.

BACKGROUND: Measurement of sodium intake in hospitalized patients is critical for their care. In this study, artificial intelligence (AI)-based imaging was performed to determine sodium intake in these patients. OBJECTIVE: The applicability of a diet management system was evaluated using AI-based imaging to assess the sodium content of diets prescribed for hospitalized patients. METHODS: Based on the information on the already investigated nutrients and quantity of food, consumed sodium was analyzed through photographs obtained before and after a meal. We used a hybrid model that first leveraged the capabilities of the You Only Look Once, version 4 (YOLOv4) architecture for the detection of food and dish areas in images. Following this initial detection, 2 distinct approaches were adopted for further classification: a custom ResNet-101 model and a hyperspectral imaging-based technique. These methodologies focused on accurate classification and estimation of the food quantity and sodium amount, respectively. The 24-hour urine sodium (UNa) value was measured as a reference for evaluating the sodium intake. RESULTS: Results were analyzed using complete data from 25 participants out of the total 54 enrolled individuals. The median sodium intake calculated by the AI algorithm (AI-Na) was determined to be 2022.7 mg per day/person (adjusted by administered fluids). A significant correlation was observed between AI-Na and 24-hour UNa, while there was a notable disparity between them. A regression analysis, considering patient characteristics (eg, gender, age, renal function, the use of diuretics, and administered fluids) yielded a formula accounting for the interaction between AI-Na and 24-hour UNa. Consequently, it was concluded that AI-Na holds clinical significance in estimating salt intake for hospitalized patients using images without the need for 24-hour UNa measurements. The degree of correlation between AI-Na and 24-hour UNa was found to vary depending on the use of diuretics. CONCLUSIONS: This study highlights the potential of AI-based imaging for determining sodium intake in hospitalized patients.

Proposal of a Network System to Solve the Problem of Small Volume in Liver Transplantation; Catholic Medical Center Network.

INTRODUCTION: Liver transplantation (LT) is a complex and demanding procedure associated with significant perioperative challenges and risks. Concerns have arisen regarding LT outcomes in low-volume centers. We implemented an integrated training and surgical team network to address these concerns within the Catholic Medical Center (CMC) network. This study presents a comprehensive review of our 9-year LT experience within the CMC medical network. METHOD: A retrospective study of LT procedures conducted between January 2013 and August 2021 in 6 CMC-affiliated hospitals was performed. One center was categorized as a high-volume center, conducting over 60 cases annually, and the remaining 5 were considered small-volume centers. The primary endpoints assessed were 1-year and 5-year survival rates. RESULTS: A total of 793 LTs were performed during the study period. The high-volume center performed 411 living donor LT (LDLT) cases and 127 deceased donor LT (DDLT) cases. Also, 146 LDLT cases and 109 DDLT cases were performed in 5 small-volume centers. One-year and 5-year patient survival for LDLT recipients was 88.3% and 78.8% in the high-volume center and 85.6% and 80.6% in the low-volume center. Five-year survival was not significantly different in small-volume centers (P = .903). For DDLT recipients, 1-year and 5-year patient survival was 80.3% and 70.6% in the high-volume center and 76.1% and 67.6% in the low-volume center. In DDLT cases, 5-year survival was not significantly different in small-volume centers (P = .445). CONCLUSION: In conclusion, comparable outcomes for liver transplantation can be obtained in a small-volume center with a high level of integrated training systems and networks.

Predicting hematoma expansion in acute spontaneous intracerebral hemorrhage: integrating clinical factors with a multitask deep learning model for non-contrast head CT.

PURPOSE: To predict hematoma growth in intracerebral hemorrhage patients by combining clinical findings with non-contrast CT imaging features analyzed through deep learning. METHODS: Three models were developed to predict hematoma expansion (HE) in 572 patients. We utilized multi-task learning for both hematoma segmentation and prediction of expansion: the Image-to-HE model processed hematoma slices, extracting features and computing a normalized DL score for HE prediction. The Clinical-to-HE model utilized multivariate logistic regression on clinical variables. The Integrated-to-HE model combined image-derived and clinical data. Significant clinical variables were selected using forward selection in logistic regression. The two models incorporating clinical variables were statistically validated. RESULTS: For hematoma detection, the diagnostic performance of the developed multi-task model was excellent (AUC, 0.99). For expansion prediction, three models were evaluated for predicting HE. The Image-to-HE model achieved an accuracy of 67.3%, sensitivity of 81.0%, specificity of 64.0%, and an AUC of 0.76. The Clinical-to-HE model registered an accuracy of 74.8%, sensitivity of 81.0%, specificity of 73.3%, and an AUC of 0.81. The Integrated-to-HE model, merging both image and clinical data, excelled with an accuracy of 81.3%, sensitivity of 76.2%, specificity of 82.6%, and an AUC of 0.83. The Integrated-to-HE model, aligning closest to the diagonal line and indicating the highest level of calibration, showcases superior performance in predicting HE outcomes among the three models. CONCLUSION: The integration of clinical findings with non-contrast CT imaging features analyzed through deep learning showed the potential for improving the prediction of HE in acute spontaneous intracerebral hemorrhage patients.

Enhanced Expression of Glycolytic Enzymes and Succinate Dehydrogenase Complex Flavoprotein Subunit A by Mesothelin Promotes Glycolysis and Mitochondrial Respiration in Myeloblasts of Acute Myeloid Leukemia.

Acute myeloid leukemia (AML) is an aggressive malignancy characterized by rapid growth and uncontrolled proliferation of undifferentiated myeloid cells. Metabolic reprogramming is commonly observed in the bone marrow of AML patients, as leukemia cells require increased ATP supply to support disease progression. In this study, we examined the potential role of mesothelin as a metabolic modulator in myeloid cells in AML. Mesothelin is a well-known marker of solid tumors that promotes cancer cell proliferation and survival. We initially analyzed alterations in mesothelin expression in the myeloblast subpopulations, defined as SSC-Alow/CD45dim, obtained from the bone marrow of AML patients using flow cytometry. Our results showed overexpression of mesothelin in 34.8% of AML patients. Subsequently, metabolic changes in leukemia cells were evaluated by comparing the oxygen consumption rates (OCR) of bone marrow samples derived from adult AML patients. Notably, a higher OCR was observed in the mesothelin-positive compared to the mesothelin-low and non-expressing groups. Treatment with recombinant human mesothelin protein enhanced OCR and increased the mRNA expression of glycolytic enzymes and mitochondrial complex II in KG1α AML cells. Notably, siRNA targeting mesothelin in KG1α cells led to the reduction of glycolysis-related gene expression but had no effect on the mitochondrial complex gene. The collective results demonstrate that mesothelin induces metabolic changes in leukemia cells, facilitating the acquisition of a rapid supply of ATP for proliferation in AML. Therefore, the targeting of mesothelin presents a potentially promising approach to mitigating the progression of AML through the inhibition of glycolysis and mitochondrial respiration in myeloid cells.

Current status of the novel food ingredient safety evaluation system.

Increasing demand for new foods, technological development, and vegan market growth have led to an increase in new food ingredients, so the need for safety assessment of these ingredients is important. Representative safety assessment systems are the Generally Recognized as Safe (GRAS) notification of the Food and Drug Administration in the USA and the novel food system of the European Food Safety Authority in the European Union. GRAS is a notification system for information on food ingredients, food additives and functional foods under the responsibility of the applicant, while the novel food system assesses the safety of food ingredients excluding food additives. In Korea, a safety evaluation system is established for temporary food ingredients, which includes food ingredients without a domestic intake history. However, safety assessment systems for novel foods from other countries and food ingredients produced by the application of new technology need to be improved.

No increased risk of osteonecrosis of the jaw in osteoporotic patients with dental implants: a nationwide cohort study.

OBJECTIVES: The occurrence of implant-associated osteonecrosis of the jaw (ONJ) has been reported in osteoporotic patients, particularly in association with bisphosphonate therapy. This study aimed to investigate the risk of implant surgery and implant presence for ONJ occurrence in osteoporotic patients longitudinally. METHODS: Based on Korean National Health Information Database, subjects over the age of 65 who were diagnosed with osteoporosis between July 2014 and December 2016 were included. The implant group included subjects who had undergone dental implant surgery between January 2017 and December 2017, while the control group included those who had no history of dental implants. The primary outcome was the occurrence of ONJ, and the date of final follow-up was December 2020. RESULTS: A total of 332,728 subjects with osteoporosis were included in the analysis: 83,182 in the implant group and 249,546 in the control group. The risk of ONJ among those who had undergone implant surgery (risk of implant surgery-associated ONJ) was not higher than that among those without implant surgery. The risk of ONJ among those with implants (risk of implant presence-associated ONJ) was lower than that among those without implants. Even in subjects with a history of bisphosphonates, steroids, periodontitis, or tooth extraction, those who had undergone implant surgery or had implants did not have a higher ONJ risk than those who had not undergone surgery or did not have implants; rather, they showed a lower risk. CONCLUSIONS: The results may suggest that dental implants are not associated with an increased risk of ONJ. A further study on whether dental implants are associated with lower ONJ risk is needed. CLINICAL RELEVANCE: Dental implants did not increase the risk of ONJ development in osteoporotic patients, even with a history of bisphosphonates. This may suggest that the risk profiles for ONJ occurrence between selective insertion of dental implants and other dentoalveolar surgery associated with infectious conditions are different.

Added prognostic value of 3D deep learning-derived features from preoperative MRI for adult-type diffuse gliomas.

BACKGROUND: To investigate the prognostic value of spatial features from whole-brain MRI using a three-dimensional (3D) convolutional neural network for adult-type diffuse gliomas. METHODS: In a retrospective, multicenter study, 1925 diffuse glioma patients were enrolled from 5 datasets: SNUH (n = 708), UPenn (n = 425), UCSF (n = 500), TCGA (n = 160), and Severance (n = 132). The SNUH and Severance datasets served as external test sets. Precontrast and postcontrast 3D T1-weighted, T2-weighted, and T2-FLAIR images were processed as multichannel 3D images. A 3D-adapted SE-ResNeXt model was trained to predict overall survival. The prognostic value of the deep learning-based prognostic index (DPI), a spatial feature-derived quantitative score, and established prognostic markers were evaluated using Cox regression. Model evaluation was performed using the concordance index (C-index) and Brier score. RESULTS: The MRI-only median DPI survival prediction model achieved C-indices of 0.709 and 0.677 (BS = 0.142 and 0.215) and survival differences (P < 0.001 and P = 0.002; log-rank test) for the SNUH and Severance datasets, respectively. Multivariate Cox analysis revealed DPI as a significant prognostic factor, independent of clinical and molecular genetic variables: hazard ratio = 0.032 and 0.036 (P < 0.001 and P = 0.004) for the SNUH and Severance datasets, respectively. Multimodal prediction models achieved higher C-indices than models using only clinical and molecular genetic variables: 0.783 vs. 0.774, P = 0.001, SNUH; 0.766 vs. 0.748, P = 0.023, Severance. CONCLUSIONS: The global morphologic feature derived from 3D CNN models using whole-brain MRI has independent prognostic value for diffuse gliomas. Combining clinical, molecular genetic, and imaging data yields the best performance.

AI-based dental caries and tooth number detection in intraoral photos: Model development and performance evaluation.

OBJECTIVES: In this study, we aimed to integrate tooth number recognition and caries detection in full intraoral photographic images using a cascade region-based deep convolutional neural network (R-CNN) model to facilitate the practical application of artificial intelligence (AI)-driven automatic caries detection in clinical practice. METHODS: Our dataset comprised 24,578 images, encompassing 4787 upper occlusal, 4347 lower occlusal, 5230 right lateral, 5010 left lateral, and 5204 frontal views. In each intraoral image, tooth numbers and, when present, dental caries, including their location and stage, were annotated using bounding boxes. A cascade R-CNN model was used for dental caries detection and tooth number recognition within intraoral images. RESULTS: For tooth number recognition, the model achieved an average mean average precision (mAP) score of 0.880. In the task of dental caries detection, the model's average mAP score was 0.769, with individual scores spanning from 0.695 to 0.893. CONCLUSIONS: The primary objective of integrating tooth number recognition and caries detection within full intraoral photographic images has been achieved by our deep learning model. The model's training on comprehensive intraoral datasets has demonstrated its potential for seamless clinical application. CLINICAL SIGNIFICANCE: This research holds clinical significance by achieving AI-driven automatic integration of tooth number recognition and caries detection in full intraoral images where multiple teeth are visible. It has the potential to promote the practical application of AI in real-life and clinical settings.