Bioinspired super-hydrophilic zwitterionic polymer armor combats thrombosis and infection of vascular catheters.

Thrombosis and infection are two major complications associated with central venous catheters (CVCs), which significantly contribute to morbidity and mortality. Antifouling coating strategies currently represent an efficient approach for addressing such complications. However, existing antifouling coatings have limitations in terms of both duration and effectiveness. Herein, we propose a durable zwitterionic polymer armor for catheters. This armor is realized by pre-coating with a robust phenol-polyamine film inspired by insect sclerotization, followed by grafting of poly-2-methacryloyloxyethyl phosphorylcholine (pMPC) via in-situ radical polymerization. The resulting pMPC coating armor exhibits super-hydrophilicity, thereby forming a highly hydrated shell that effectively prevents bacterial adhesion and inhibits the adsorption and activation of fibrinogen and platelets in vitro. In practical applications, the armored catheters significantly reduced inflammation and prevented biofilm formation in a rat subcutaneous infection model, as well as inhibited thrombus formation in a rabbit jugular vein model. Overall, our robust zwitterionic polymer coating presents a promising solution for reducing infections and thrombosis associated with vascular catheters.

Active Dual-Protein Coating Assisted by Stepwise Protein-Protein Interactions Assembly Reduces Thrombosis and Infection.

Universal protein coatings have recently gained wide interest in medical applications due to their biocompatibility and ease of fabrication. However, the challenge persists in protein activity preservation, significantly complicating the functional design of these coatings. Herein, an active dual-protein surface engineering strategy assisted by a facile stepwise protein-protein interactions assembly (SPPIA) method for catheters to reduce clot formation and infection is proposed. This strategy is realized first by the partial oxidation of bovine serum albumin (BSA) and lysozyme (LZM) for creating stable nucleation platforms via hydrophobic interaction, followed by the assembly of nonoxidized BSA (pI, the isoelectric point, ≈4.7) and LZM (pI ≈11) through electrostatic interaction owing to their opposite charge under neutral conditions. The SPPIA method effectively preserves the conformation and functionality of both BSA and LZM, thus endowing the resultant coating with potent antithrombotic and bactericidal properties. Furthermore, the stable nucleation platform ensures the adhesion and durability of the coating, resisting thrombosis and bacterial proliferation even after 15 days of PBS immersion. Overall, the SPPIA approach not only provides a new strategy for the fabrication of active protein coatings but also shows promise for the surface engineering technology of catheters.

Multi-center application of a convolutional neural network for preoperative detection of cavernous sinus invasion in pituitary adenomas.

OBJECTIVE: Cavernous sinus invasion (CSI) plays a pivotal role in determining management in pituitary adenomas. The study aimed to develop a Convolutional Neural Network (CNN) model to diagnose CSI in multiple centers. METHODS: A total of 729 cases were retrospectively obtained in five medical centers with (n = 543) or without CSI (n = 186) from January 2011 to December 2021. The CNN model was trained using T1-enhanced MRI from two pituitary centers of excellence (n = 647). The other three municipal centers (n = 82) as the external testing set were imported to evaluate the model performance. The area-under-the-receiver-operating-characteristic-curve values (AUC-ROC) analyses were employed to evaluate predicted performance. Gradient-weighted class activation mapping (Grad-CAM) was used to determine models' regions of interest. RESULTS: The CNN model achieved high diagnostic accuracy (0.89) in identifying CSI in the external testing set, with an AUC-ROC value of 0.92 (95% CI, 0.88-0.97), better than CSI clinical predictor of diameter (AUC-ROC: 0.75), length (AUC-ROC: 0.80), and the three kinds of dichotomizations of the Knosp grading system (AUC-ROC: 0.70-0.82). In cases with Knosp grade 3A (n = 24, CSI rate, 0.35), the accuracy the model accounted for 0.78, with sensitivity and specificity values of 0.72 and 0.78, respectively. According to the Grad-CAM results, the views of the model were confirmed around the sellar region with CSI. CONCLUSIONS: The deep learning model is capable of accurately identifying CSI and satisfactorily able to localize CSI in multicenters.

An insect sclerotization-inspired antifouling armor on biomedical devices combats thrombosis and embedding.

Thrombus formation and tissue embedding significantly impair the clinical efficacy and retrievability of temporary interventional medical devices. Herein, we report an insect sclerotization-inspired antifouling armor for tailoring temporary interventional devices with durable resistance to protein adsorption and the following protein-mediated complications. By mimicking the phenol-polyamine chemistry assisted by phenol oxidases during sclerotization, we develop a facile one-step method to crosslink bovine serum albumin (BSA) with oxidized hydrocaffeic acid (HCA), resulting in a stable and universal BSA@HCA armor. Furthermore, the surface of the BSA@HCA armor, enriched with carboxyl groups, supports the secondary grafting of polyethylene glycol (PEG), further enhancing both its antifouling performance and durability. The synergy of robustly immobilized BSA and covalently grafted PEG provide potent resistance to the adhesion of proteins, platelets, and vascular cells in vitro. In ex vivo blood circulation experiment, the armored surface reduces thrombus formation by 95 %. Moreover, the antifouling armor retained over 60 % of its fouling resistance after 28 days of immersion in PBS. Overall, our armor engineering strategy presents a promising solution for enhancing the antifouling properties and clinical performance of temporary interventional medical devices.

Glycocalyx-inspired dynamic antifouling surfaces for temporary intravascular devices.

Protein and cell adhesion on temporary intravascular devices can lead to thrombosis and tissue embedment, significantly increasing complications and device retrieval difficulties. Here, we propose an endothelial glycocalyx-inspired dynamic antifouling surface strategy for indwelling catheters and retrievable vascular filters to prevent thrombosis and suppress intimal embedment. This strategy is realized on the surfaces of substrates by the intensely dense grafting of hydrolyzable endothelial polysaccharide hyaluronic acid (HA), assisted by an amine-rich phenol-polyamine universal platform. The resultant super-hydrophilic surface exhibits potent antifouling property against proteins and cells. Additionally, the HA hydrolysis induces continuous degradation of the coating, enabling removal of inevitable biofouling on the surface. Moreover, the dense grafting of HA also ensures the medium-term effectiveness of this dynamic antifouling surface. The coated catheters maintain a superior anti-thrombosis capacity in ex vivo blood circulation after 30 days immersion. In the abdominal veins of rats, the coated implants show inhibitory effects on intimal embedment up to 2 months. Overall, we envision that this glycocalyx-inspired dynamic antifouling surface strategy could be a promising surface engineering technology for temporary intravascular devices.

Neurodevelopment effects of early life bisphenol-A exposure on visual memory: Insights into recovery dynamics.

Bisphenol A (BPA), a ubiquitous endocrine disruptor, is implicated in the cognitive deficits observed in both children and animals. Especially, BPA-induced spatial memory deterioration during the whole development phase of rodents has been well delineated. However, whether BPA exposure on the different development phases exerts similar effects on the prefrontal cortex (PFC) dependent visual memory is still elusive. Here, we chose two exposure windows, the whole gestation and lactation phases (E0∼P21) and the whole juvenile and adolescent phases (P22∼P60), for exposing rats to BPA. The visual memory of those rats was accessed by object recognition testing in the open field after BPA exposure and a constant recovery interval. The results revealed a substantial decline of visual memory under both exposure conditions, accompanied by an increase in anxiety-like behavior in BPA-exposed rats. Notably, after a 20-day recovery period, those behavioral changes induced by BPA exposure during P22∼60, not E0∼P21, were reversed compared to the control rats. According to morphological analysis of those rats after recovery, we found that the spine density of pyramidal neurons in the PFC were significant decreased in rats with BPA exposure during E0∼P21 and there was no difference between rats with or without BPA exposure during P22∼P60. Additionally, a similar change trend in excitatory receptors expression was observed under both exposure conditions. After an additional 20 days of recovery, the behavioral changes in rats with perinatal BPA exposure reverted to the normal status. Our present findings illuminate the dynamic effects of BPA on PFC-dependent functions across two crucial early developmental stages of life.

Machine Learning Approaches to Differentiate Sellar-Suprasellar Cystic Lesions on Magnetic Resonance Imaging.

Cystic lesions are common lesions of the sellar region with various pathological types, including pituitary apoplexy, Rathke's cleft cyst, cystic craniopharyngioma, etc. Suggested surgical approaches are not unique when dealing with different cystic lesions. However, cystic lesions with different pathological types were hard to differentiate on MRI with the naked eye by doctors. This study aimed to distinguish different pathological types of cystic lesions in the sellar region using preoperative magnetic resonance imaging (MRI). Radiomics and deep learning approaches were used to extract features from gadolinium-enhanced MRIs of 399 patients enrolled at Peking Union Medical College Hospital over the past 15 years. Paired imaging differentiations were performed on four subtypes, including pituitary apoplexy, cystic pituitary adenoma (cysticA), Rathke's cleft cyst, and cystic craniopharyngioma. Results showed that the model achieved an average AUC value of 0.7685. The model based on a support vector machine could distinguish cystic craniopharyngioma from Rathke's cleft cyst with the highest AUC value of 0.8584. However, distinguishing cystic apoplexy from pituitary apoplexy was difficult and almost unclassifiable with any algorithms on any feature set, with the AUC value being only 0.6641. Finally, the proposed methods achieved an average Accuracy of 0.7532, which outperformed the traditional clinical knowledge-based method by about 8%. Therefore, in this study, we first fill the gap in the existing literature and provide a non-invasive method for accurately differentiating between these lesions, which could improve preoperative diagnosis accuracy and help to make surgery plans in clinical work.

Corrosion-tailoring, osteogenic, anti-inflammatory, and antibacterial aspirin-loaded organometallic hydrogel composite coating on biodegradable Zn for orthopedic applications.

Zn and its alloys are receiving increasing interest for biodegradable orthopedic implant applications owing to their moderate corrosion rate and the potential functionality of Zn2+. However, their non-uniform corrosion behavior and insufficient osteogenic, anti-inflammatory, and antibacterial properties do not meet the comprehensive requirements of orthopedic implants in clinical use. Herein, an aspirin (an acetylsalicylic acid, ASA, 10, 50, 100, and 500 mg/L)-loaded carboxymethyl chitosan (CMC)/gelatin (Gel)-Zn2+ organometallic hydrogel composite coating (CMC/Gel&Zn2+/ASA) was fabricated on a Zn surface via an alternating dip-coating method, aiming to obtain a material with these comprehensive properties improved. The organometallic hydrogel composite coatings, ca. 12-16 µm in thickness, showed compact, homogeneous, and micro-bulge structured surface morphology. The coatings protected well the Zn substrate from pitting/localized corrosion and contained the release of the bioactive components, Zn2+ and ASA, in a sustained and stable manner in long-term in vitro immersions in Hank's solution. The coated Zn showed greater ability to promote proliferation and osteogenic differentiation for MC3T3-E1 osteoblasts, and better anti-inflammatory capacity when compared with uncoated Zn. Additionally, this coating displayed excellent antibacterial activity against both Escherichia coli (>99 % antibacterial rate) and Staphylococcus aureus (>98 % antibacterial rate). Such appealing properties can be attributed to the compositional nature of the coating, namely the sustained release of Zn2+ and ASA, as well as the surface physiochemical properties because of its unique microstructure. This organometallic hydrogel composite coating can be considered a promising option for the surface modification of biodegradable Zn-based orthopedic implants among others.

Metal-organic Zn-zoledronic acid and 1-hydroxyethylidene-1,1-diphosphonic acid nanostick-mediated zinc phosphate hybrid coating on biodegradable Zn for osteoporotic fracture healing implants.

Zn and its alloys are increasingly under consideration for biodegradable bone fracture fixation implants owing to their attractive biodegradability and mechanical properties. However, their clinical application is a challenge for osteoporotic bone fracture healing, due to their uneven degradation mode, burst release of zinc ions, and insufficient osteo-promotion and osteo-resorption regulating properties. In this study, a type of Zn2+ coordinated zoledronic acid (ZA) and 1-hydroxyethylidene-1,1-diphosphonic acid (HEDP) metal-organic hybrid nanostick was synthesized, which was further mixed into zinc phosphate (ZnP) solution to mediate the deposition and growth of ZnP to form a well-integrated micro-patterned metal-organic/inorganic hybrid coating on Zn. The coating protected noticeably the Zn substrate from corrosion, in particular reducing its localized occurrence as well as suppressing its Zn2+ release. Moreover, the modified Zn was osteo-compatible and osteo-promotive and, more important, performed osteogenesis in vitro and in vivo of well-balanced pro-osteoblast and anti-osteoclast responses. Such favorable functionalities are related to the nature of its bioactive components, especially the bio-functional ZA and the Zn ions it contains, as well as its unique micro- and nano-scale structure. This strategy provides not only a new avenue for surface modification of biodegradable metals but also sheds light on advanced biomaterials for osteoporotic fracture and other applications. STATEMENT OF SIGNIFICANCE: Developing appropriate biodegradable metallic materials is of clinical relevance for osteoporosis fracture healing, whereas current strategies are short of good balance between the bone formation and resorption. Here, we designed a micropatterned metal-organic nanostick mediated zinc phosphate hybrid coating modified Zn biodegradable metal to fulfill such a balanced osteogenicity. The in vitro assays verified the coated Zn demonstrated outstanding pro-osteoblasts and anti-osteoclasts properties and the coated intramedullary nail promoted fracture healing well in an osteoporotic femur fracture rat model. Our strategy may offer not only a new avenue for surface modification of biodegradable metals but also shed light on better understanding of new advanced biomaterials for orthopedic application among others.

Responses of biomass accumulation and nutrient utilization along a phosphorus supply gradient in Leymus chinensis.

Phosphorus (P) deficiencies are widespread in calcareous soils. The poor availability of nitrogen (N) and P in soils often restricts crop growth. However, the effects of P addition on plant growth and plant nutrient transport changes during the establishment of Leymus chinensis fields in Xinjiang are not clear. We investigated the responses of Leymus chinensis biomass and nutrient absorption and utilization to changes in soil N and P by adding P (0, 15.3, 30.6, and 45.9 kg P ha-1 year-1) with basally applied N fertilizer (150 kg N ha-1 year-1). The results showed that (a) Principal component analysis (PCA) of biomass, nutrient accumulation, soil available P, and soil available N during the different periods of Leymus chinensis growth showed that their cumulative contributions during the jointing and harvest periods reached 95.4% and 88%, respectively. (b) Phosphorus use efficiency (PUE) increased with the increase of P fertilizer gradient and then decreased and the maximum PUE was 13.14% under moderate P addition. The accumulation of biomass and nutrients in Leymus chinensis can be effectively improved by the addition of P fertilizer at 30.6 kg ha-1. Different P additions either moderately promoted or excessively inhibited Leymus chinensis growth and nutrient utilization.

Thyroxine changes in COVID-19 pandemic: A systematic review and meta-analysis.

Objective: COVID-19 infection may affect thyroid function. However, changes in thyroid function in COVID-19 patients have not been well described. This systematic review and meta-analysis assess thyroxine levels in COVID-19 patients, compared with non-COVID-19 pneumonia and healthy cohorts during the COVID-19 epidemic. Methods: A search was performed in English and Chinese databases from inception to August 1, 2022. The primary analysis assessed thyroid function in COVID-19 patients, comparing non-COVID-19 pneumonia and healthy cohorts. Secondary outcomes included different severity and prognoses of COVID-19 patients. Results: A total of 5873 patients were enrolled in the study. The pooled estimates of TSH and FT3 were significantly lower in patients with COVID-19 and non-COVID-19 pneumonia than in the healthy cohort (P < 0.001), whereas FT4 were significantly higher (P < 0.001). Patients with the non-severe COVID-19 showed significant higher in TSH levels than the severe (I2 = 89.9%, P = 0.002) and FT3 (I2 = 91.9%, P < 0.001). Standard mean differences (SMD) of TSH, FT3, and FT4 levels of survivors and non-survivors were 0.29 (P= 0.006), 1.11 (P < 0.001), and 0.22 (P < 0.001). For ICU patients, the survivors had significantly higher FT4 (SMD=0.47, P=0.003) and FT3 (SMD=0.51, P=0.001) than non-survivors. Conclusions: Compared with the healthy cohort, COVID-19 patients showed decreased TSH and FT3 and increased FT4, similar to non-COVID-19 pneumonia. Thyroid function changes were related to the severity of COVID-19. Thyroxine levels have clinical significance for prognosis evaluation, especially FT3.

A "built-up" composite film with synergistic functionalities on Mg-2Zn-1Mn bioresorbable stents improves corrosion control effects and biocompatibility.

Control of premature corrosion of magnesium (Mg) alloy bioresorbable stents (BRS) is frequently achieved by the addition of rare earth elements. However, limited long-term experience with these elements causes concerns for clinical application and alternative methods of corrosion control are sought after. Herein, we report a "built-up" composite film consisting of a bottom layer of MgF2 conversion coating, a sandwich layer of a poly (1, 3-trimethylene carbonate) (PTMC) and 3-aminopropyl triethoxysilane (APTES) co-spray coating (PA) and on top a layer of poly (lactic-co-glycolic acid) (PLGA) ultrasonic spray coating to decorate the rare earth element-free Mg-2Zn-1Mn (ZM21) BRS for tailoring both corrosion resistance and biological functions. The developed "built-up" composite film shows synergistic functionalities, allowing the compression and expansion of the coated ZM21 BRS on an angioplasty balloon without cracking or peeling. Of special importance is that the synergistic corrosion control effects of the "built-up" composite film allow for maintaining the mechanical integrity of stents for up to 3 months, where complete biodegradation and no foreign matter residue were observed about half a year after implantation in rabbit iliac arteries. Moreover, the functionalized ZM21 BRS accomplished re-endothelialization within one month.

Predicting delayed remission in Cushing's disease using radiomics models: a multi-center study.

Purpose: No multi-center radiomics models have been built to predict delayed remission (DR) after transsphenoidal surgery (TSS) in Cushing's disease (CD). The present study aims to build clinical and radiomics models based on data from three centers to predict DR after TSS in CD. Methods: A total of 122 CD patients from Peking Union Medical College Hospital, Xuanwu Hospital, and Fuzhou General Hospital were enrolled between January 2000 and January 2019. The T1-weighted gadolinium-enhanced MRI images and clinical data were used as inputs to build clinical and radiomics models. The regions of interest (ROI) of MRI images were automatically defined by a deep learning algorithm developed by our team. The area under the curve (AUC) of receiver operating characteristic (ROC) curves was used to evaluate the performance of the models. In total, 10 machine learning algorithms were used to construct models. Results: The overall DR rate is 44.3% (54/122). According to multivariate Logistic regression analysis, patients with higher BMI and lower postoperative cortisol levels are more likely to achieve a higher rate of delayed remission. Among the 10 models, XGBoost achieved the best performance among all models in both clinical and radiomics models with AUC values of 0.767 and 0.819 respectively. The results from SHAP value and LIME algorithms revealed that postoperative cortisol level (PoC) and BMI were the most important features associated with DR. Conclusion: Radiomics models can be built as an effective noninvasive method to predict DR and might be useful in assisting neurosurgeons in making therapeutic plans after TSS for CD patients. These results are preliminary and further validation in a larger patient sample is needed.

Osteogenic and angiogenic bioactive collagen entrapped calcium/zinc phosphates coating on biodegradable Zn for orthopedic implant applications.

Zinc is becoming one of the leading candidate materials for biodegradable orthopedic implants owing to its attractive properties in terms of degradation behavior and mechanical properties. However, the insufficient surface bio-activities postpone its clinical application. In this study, an organic-inorganic collagen entrapped calcium/zinc phosphates coating was constructed on Zn surface to lessen Zn2+ releasing rate and to leverage the surface osteogenic and angiogenic properties. Collagen molecules were immobilized onto Zn substrate and subsequently coordinated with calcium and zinc ions to promote the CaZnP inorganic phase growth, ensuing an intertwined collagen-CaZnP hybrid system. Consequently, the hybrid coating was highly coalesced and compact. Such high quality warranted the contained Zn2+ releasing in a tolerable rate favorable for cells viability. The collagen-CaZnP coated Zn showed remarkedly stronger osteogenicity as compared to the untreated Zn, ascertained by the MC3T3-E1 osteoblast cell proliferation and differentiation assays, such as alkaline phosphatase expression and calcium nodule formation results. In addition, this hybrid coating supported human umbilical vein endothelial cells (HUVECs) migration and tube formation. The enhanced osteogenic and angiogenic properties could be ascribed to the nature of collagen and calcium/zinc phosphate components, the hybrid micro/nano-structure as well as the ability of controlling the Zn2+ release of Zn substrate into a suitable concentration range. Our strategy provides a new avenue to surface modification of biodegradable metals for bone regenerative perspective.

Biodegradable Zn-Cu-Fe Alloy as a Promising Material for Craniomaxillofacial Implants: An in vitro Investigation into Degradation Behavior, Cytotoxicity, and Hemocompatibility.

Zinc-based nanoparticles, nanoscale metal frameworks and metals have been considered as biocompatible materials for bone tissue engineering. Among them, zinc-based metals are recognized as promising biodegradable materials thanks to their moderate degradation rate ranging between magnesium and iron. Nonetheless, materials' biodegradability and the related biological response depend on the specific implant site. The present study evaluated the biodegradability, cytocompatibility, and hemocompatibility of a hot-extruded zinc-copper-iron (Zn-Cu-Fe) alloy as a potential biomaterial for craniomaxillofacial implants. Firstly, the effect of fetal bovine serum (FBS) on in vitro degradation behavior was evaluated. Furthermore, an extract test was used to evaluate the cytotoxicity of the alloy. Also, the hemocompatibility evaluation was carried out by a modified Chandler-Loop model. The results showed decreased degradation rates of the Zn-Cu-Fe alloy after incorporating FBS into the medium. Also, the alloy exhibited acceptable toxicity towards RAW264.7, HUVEC, and MC3T3-E1 cells. Regarding hemocompatibility, the alloy did not significantly alter erythrocyte, platelet, and leukocyte counts, while the coagulation and complement systems were activated. This study demonstrated the predictable in vitro degradation behavior, acceptable cytotoxicity, and appropriate hemocompatibility of Zn-Cu-Fe alloy; therefore, it might be a candidate biomaterial for craniomaxillofacial implants.

Mussel-Inspired and Bioclickable Peptide Engineered Surface to Combat Thrombosis and Infection.

Thrombosis and infections are the two major complications associated with extracorporeal circuits and indwelling medical devices, leading to significant mortality in clinic. To address this issue, here, we report a biomimetic surface engineering strategy by the integration of mussel-inspired adhesive peptide, with bio-orthogonal click chemistry, to tailor the surface functionalities of tubing and catheters. Inspired by mussel adhesive foot protein, a bioclickable peptide mimic (DOPA)4-azide-based structure is designed and grafted on an aminated tubing robustly based on catechol-amine chemistry. Then, the dibenzylcyclooctyne (DBCO) modified nitric oxide generating species of 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) chelated copper ions and the DBCO-modified antimicrobial peptide (DBCO-AMP) are clicked onto the grafted surfaces via bio-orthogonal reaction. The combination of the robustly grafted AMP and Cu-DOTA endows the modified tubing with durable antimicrobial properties and ability in long-term catalytically generating NO from endogenous s-nitrosothiols to resist adhesion/activation of platelets, thus preventing the formation of thrombosis. Overall, this biomimetic surface engineering technology provides a promising solution for multicomponent surface functionalization and the surface bioengineering of biomedical devices with enhanced clinical performance.

Application of Convolutional Neural Network in the Diagnosis of Cavernous Sinus Invasion in Pituitary Adenoma.

Objectives: Convolutional neural network (CNN) is a deep-learning method for image classification and recognition based on a multi-layer NN. In this study, CNN was used to accurately assess cavernous sinus invasion (CSI) in pituitary adenoma (PA). Methods: A total of 371 patients with PA were enrolled in the retrospective study. The cohort was divided into the invasive (n = 102) and non-invasive groups (n = 269) based on surgically confirmed CSI. Images were selected on the T1-enhanced imaging on MR scans. The cohort underwent a fivefold division of randomized datasets for cross-validation. Then, a tenfold augmented dataset (horizontal flip and rotation) of the training set was enrolled in the pre-trained Resnet50 model for transfer learning. The testing set was imported into the trained model for evaluation. Gradient-weighted class activation mapping (Grad-CAM) was used to obtain the occlusion map. The diagnostic values were compared with different dichotomizations of the Knosp grading system (grades 0-1/2-4, 0-2/3a-4, and 0-3a/3b-4). Results: Based on Knosp grades, 20 cases of grade 0, 107 cases of grade 1, 82 cases of grade 2, 104 cases of grade 3a, 22 cases of grade 3b, and 36 cases of grade 4 were recorded. The CSI rates were 0%, 3.7%, 18.3%, 37.5%, 54.5%, and 88.9%. The predicted accuracies of the three dichotomies were 60%, 74%, and 81%. The area under the receiver operating characteristic (AUC-ROC) of Knosp grade for CSI prediction was 0.84; the cutoff was 2.5 with a Youden value of 0.62. The accuracies of the CNN model ranged from 0.80 to 0.96, with AUC-ROC values ranging from 0.89 to 0.98. The Grad-CAM saliency maps confirmed that the region of interest of the model was around the sellar region. Conclusions: We constructed a CNN model with a high proficiency at CSI diagnosis. A more accurate CSI identification was achieved with the constructed CNN than the Knosp grading system.

Diagnosis, Manifestations, Laboratory Investigations, and Prognosis in Pediatric and Adult Cushing's Disease in a Large Center in China.

Purpose: Cushing's disease (CD) is a rare disease that contributes to 70-80% hypercortisolemia, which presents similarities and differences between pediatric and adult patients, and even between male and female patients. However, the comparative study of CD between different age groups and different genders is still insufficient. The aim of the study is to make a systematic comparison to reveal the gender differences in children and adult patients of CD, helping clinicians to provide optimal treatment for different groups of patients. Methods: We conducted a retrospective research consisting of 30 pediatric and 392 adult CD patients in a single center in Peking Union Medical College Hospital. All 422 patients showed symptoms related to hypercortisolism and received adenoma excision surgery in the department of neurosurgery between 2014 and 2020. Results: For the accuracy of diagnosis, the sensitivity of BIPSS at baseline in pediatric patients was lower than in adults (75 vs. 91%, P = 0.054) but increased greatly after desmopressin stimulation (94 vs. 95%). However, the accuracy of lateralization for BIPSS was not preferred for prediction. As for clinical manifestations, growth retardation, weight gain, hirsutism, and acne were more prevalent for children, while for adults, hypertension, osteopenia, glucometabolic disorder, easy bruising, hair loss, and weight loss were more frequently seen. As previously reported, we observed a significant difference between the male prevalence of pediatric and adult patients (50 vs. 17%, P < 0.001), which was possibly caused by the more severe and earlier onset of a series of symptoms. Gender-related comparison showed greater morbidity of nephrolithiasis, hypokalemia, hypertension, easy bruising, osteopenia, and striae for male patients, while irregular menses, hirsutism, and hair loss were more common for female patients. Further analysis showed that the secretory activity of the PA axis was higher for males, presenting as the more remarkable alteration of laboratory parameters and contributing to the more severe clinical manifestations. For patients treated with transsphenoidal pituitary surgery (TSS), the immediate prognosis could be predicted by operation history, invasiveness, Ki-67, and information provided by MRI, including tumor size and Knosp grading. However, we still lack methods to predict long-term prognosis. Conclusions: Our study is the first detailed and systematic comparison between pediatric and adult CD patients. Further exploration of the impact of CD on different genders reveals a more severe and probably an earlier-onset pattern of CD for male patients.

Electronic Medical Records as Input to Predict Postoperative Immediate Remission of Cushing's Disease: Application of Word Embedding.

BACKGROUND: No existing machine learning (ML)-based models use free text from electronic medical records (EMR) as input to predict immediate remission (IR) of Cushing's disease (CD) after transsphenoidal surgery. PURPOSE: The aim of the present study is to develop an ML-based model that uses EMR that include both structured features and free text as input to preoperatively predict IR after transsphenoidal surgery. METHODS: A total of 419 patients with CD from Peking Union Medical College Hospital were enrolled between January 2014 and August 2020. The EMR of the patients were embedded and transformed into low-dimensional dense vectors that can be included in four ML-based models together with structured features. The area under the curve (AUC) of receiver operating characteristic curves was used to evaluate the performance of the models. RESULTS: The overall remission rate of the 419 patients was 75.7%. From the results of logistic multivariate analysis, operation (p < 0.001), invasion of cavernous sinus from MRI (p = 0.046), and ACTH (p = 0.024) were strongly correlated with IR. The AUC values for the four ML-based models ranged from 0.686 to 0.793. The highest AUC value (0.793) was for logistic regression when 11 structured features and "individual conclusions of the case by doctor" were included. CONCLUSION: An ML-based model was developed using both structured and unstructured features (after being processed using a word embedding method) as input to preoperatively predict postoperative IR.

Machine-Learning Prediction of Postoperative Pituitary Hormonal Outcomes in Nonfunctioning Pituitary Adenomas: A Multicenter Study.

Objective: No accurate predictive models were identified for hormonal prognosis in non-functioning pituitary adenoma (NFPA). This study aimed to develop machine learning (ML) models to facilitate the prognostic assessment of pituitary hormonal outcomes after surgery. Methods: A total of 215 male patients with NFPA, who underwent surgery in four medical centers from 2015 to 2021, were retrospectively reviewed. The data were pooled after heterogeneity assessment, and they were randomly divided into training and testing sets (172:43). Six ML models and logistic regression models were developed using six anterior pituitary hormones. Results: Only thyroid-stimulating hormone (p < 0.001), follicle-stimulating hormone (p < 0.001), and prolactin (PRL; p < 0.001) decreased significantly following surgery, whereas growth hormone (GH) (p < 0.001) increased significantly. The postoperative GH (p = 0.07) levels were slightly higher in patients with gross total resection, but the PRL (p = 0.03) level was significantly lower than that in patients with subtotal resection. The optimal model achieved area-under-the-receiver-operating-characteristic-curve values of 0.82, 0.74, and 0.85 in predicting hormonal hypofunction, new deficiency, and hormonal recovery following surgery, respectively. According to feature importance analyses, the preoperative levels of the same type and other hormones were all important in predicting postoperative individual hormonal hypofunction. Conclusion: Fluctuation in anterior pituitary hormones varies with increases and decreases because of transsphenoidal surgery. The ML models could accurately predict postoperative pituitary outcomes based on preoperative anterior pituitary hormones in NFPA.