Vitamin D3 attenuates autoimmune thyroiditis by regulating Th17/Treg cell differentiation via YAP/JAK1/STAT1 axis.

BACKGROUND: Autoimmune thyroiditis (AITD) is an organ-specific autoimmune disease. Substantial evidence suggests that Vitamin D (VitD) deficiency is closely associated with an increased risk of AITD. However, the effects of VitD3 on immune cells, especially Th17/Treg cell subsets, and the underlying molecular mechanism in AITD have not yet been investigated. METHODS: An experimental autoimmune thyroiditis (EAT) mouse model was established with a high-iodine diet. After 8 weeks, thyroid injury was assessed using hematoxylin and eosin (H&E) staining. ELISA was employed to measure serum levels of thyroxine (T3 and T4), thyroid autoimmune antibodies (Tg-Ab and TPO-Ab), and inflammatory cytokines. Flow cytometry and multiplex fluorescence immunohistochemical (mIHC) assays were used to analyze Th17/Treg cell subsets. The CCK-8 and flow cytometry assays were used to determine cell viability and apoptosis. RESULTS: Administration of VitD3 reduced thyroid follicle destruction, decreased lymphocyte infiltration, and lowered T3, T4, Tg-Ab, and TPO-Ab serum levels in EAT mice. VitD3 treatment also reduced the frequency of Th17 cells while promoting the Treg cell subset both in the thyroid tissue and in the splenocytes cultured in vitro. Furthermore, VitD3 administration suppressed the production of inflammatory cytokines in EAT mice. VitD3 was also found to regulate Treg cells' differentiation, viability, and apoptosis. Mechanistically, we discovered that VitD3 treatment upregulated YAP expression and activated the JAK/STAT pathway. Rescue assays confirmed that depletion of YAP counteracted the effects of VitD3 on Treg cell differentiation and function. CONCLUSION: Vitamin D3 attenuates AITD by modulating Th17/Treg cell balance via regulating the YAP/JAK1/STAT1 axis.

KAT7 Suppresses Tumorigenesis in Clear Cell Renal Cell Carcinoma (ccRCC) by Regulating Cell Cycle and Ferroptosis Sensitivity.

Clear cell renal cell carcinoma (ccRCC) is one of the most aggressive malignancies in the urological system, known for its high immunogenicity. However, its pathogenesis remains unclear. This study utilized bioinformatics algorithms and in vitro experiments to investigate the role of KAT7 in ccRCC. The results indicate that KAT7 is significantly downregulated in ccRCC tissues and cell lines, which is linked to distant metastasis and unfavorable outcomes in ccRCC patients. Overexpression of KAT7 in vitro notably decreased the proliferation, migration, and invasion of renal cancer cells and inhibited Epithelial-Mesenchymal Transition (EMT). Additionally, Gene Set Enrichment Analysis (GSEA) demonstrated that KAT7-related gene functions are associated with cell cycle and ferroptosis transcription factors. Treatment with a KAT7 acetylation inhibitor in ccRCC cell lines reversed the S phase arrest caused by KAT7 overexpression. Similarly, ferroptosis inhibitors alleviated ferroptosis induced by overexpressed KAT7. In conclusion, the findings suggest that KAT7 acts as a tumor suppressor in ccRCC by modulating the cell cycle and ferroptosis sensitivity, underscoring its potential as a therapeutic target and prognostic biomarker for renal cell carcinoma patients.

NBS1 lactylation is required for efficient DNA repair and chemotherapy resistance.

The Warburg effect is a hallmark of cancer that refers to the preference of cancer cells to metabolize glucose anaerobically rather than aerobically1,2. This results in substantial accumulation of lacate, the end product of anaerobic glycolysis, in cancer cells3. However, how cancer metabolism affects chemotherapy response and DNA repair in general remains incompletely understood. Here we report that lactate-driven lactylation of NBS1 promotes homologous recombination (HR)-mediated DNA repair. Lactylation of NBS1 at lysine 388 (K388) is essential for MRE11-RAD50-NBS1 (MRN) complex formation and the accumulation of HR repair proteins at the sites of DNA double-strand breaks. Furthermore, we identify TIP60 as the NBS1 lysine lactyltransferase and the 'writer' of NBS1 K388 lactylation, and HDAC3 as the NBS1 de-lactylase. High levels of NBS1 K388 lactylation predict poor patient outcome of neoadjuvant chemotherapy, and lactate reduction using either genetic depletion of lactate dehydrogenase A (LDHA) or stiripentol, a lactate dehydrogenase A inhibitor used clinically for anti-epileptic treatment, inhibited NBS1 K388 lactylation, decreased DNA repair efficacy and overcame resistance to chemotherapy. In summary, our work identifies NBS1 lactylation as a critical mechanism for genome stability that contributes to chemotherapy resistance and identifies inhibition of lactate production as a promising therapeutic cancer strategy.

Classification of variant portal vein anatomy based on three-dimensional CT: surgical implications.

PURPOSES: The purpose of this study was to develop a new and more comprehensive classification system for portal vein (PV) variations using three-dimensional visualization and evaluation (3DVE) and to discuss the prevalence rates and clinical implications of the variants. METHODS: The anatomies of PVs were tracked and analyzed by using three-dimensional visualization of CT images acquired between 2013 and 2022. Scans from 200 adults were evaluated and a total of 178 patients (N = 178) were included in the study. The new classification system, named BLB classification, was developed based on the level of the absent PV branch in each variant anatomy. RESULTS: Using the BLB classification system, PVs were divided into thirteen subtypes. Only 82.6-84.8% of the portal veins of the 178 patients were depicted in Atri's, Cheng's or Covey's classification, compared with 100% identified by the BLB classification. The BLB classification was validated against external data sets from previous studies, with 97.0-98.9% of patients classified by the BLB system. CONCLUSION: Variant PV anatomies are more commonly seen based on 3DVE than in previous reports. The BLB classification covers almost all portal vein variants and may be used for planning liver surgery.

A unified model-based framework for doublet or multiplet detection in single-cell multiomics data.

Droplet-based single-cell sequencing techniques rely on the fundamental assumption that each droplet encapsulates a single cell, enabling individual cell omics profiling. However, the inevitable issue of multiplets, where two or more cells are encapsulated within a single droplet, can lead to spurious cell type annotations and obscure true biological findings. The issue of multiplets is exacerbated in single-cell multiomics settings, where integrating cross-modality information for clustering can inadvertently promote the aggregation of multiplet clusters and increase the risk of erroneous cell type annotations. Here, we propose a compound Poisson model-based framework for multiplet detection in single-cell multiomics data. Leveraging experimental cell hashing results as the ground truth for multiplet status, we conducted trimodal DOGMA-seq experiments and generated 17 benchmarking datasets from two tissues, involving a total of 280,123 droplets. We demonstrated that the proposed method is an essential tool for integrating cross-modality multiplet signals, effectively eliminating multiplet clusters in single-cell multiomics data-a task at which the benchmarked single-omics methods proved inadequate.

Targeting Ferroptosis as an Advance Strategy in Cancer Therapy.

Ferroptosis, triggered by the buildup of lipid peroxidation and reliance on iron, is crucial in maintaining cellular balance. Research related to ferroptosis has surged due to its distinct characteristics compared to other forms of controlled cell death, both in terms of mechanisms and appearance. Scientists believe that directing efforts towards targeting ferroptosis could pave the way for innovative precision cancer treatments, addressing challenges such as cancer recurrence and resistance. This review systematically outlines the molecular mechanisms behind ferroptosis, the substances that induce ferroptosis, and how different cancers respond to ferroptotic inducers. Also, the study further looks into the molecular basis of ferroptosis in tumor biology. It provides a conceptual framework illustrating its interaction with the tumor immune microenvironment, guiding treatment choices, predicting efficacy, exploring combination therapies, and identifying new therapeutic targets to overcome cancer resistance to standard treatments. Finally, it highlights key issues and obstacles in future research and clinical translation of ferroptosis as a potential strategy in cancer therapy.

Functional modification of gut bacteria for disease diagnosis and treatment.

Intestinal bacteria help keep humans healthy by regulating lipid and glucose metabolism as well as the immunological and neurological systems. Oral treatment using intestinal bacteria is limited by the high acidity of stomach fluids and the immune system's attack on foreign bacteria. Scientists have created coatings and workarounds to overcome these limitations and improve bacterial therapy. These preparations have demonstrated promising outcomes, with advances in synthetic biology and optogenetics improving their focused colonization and controlled release. Engineering bacteria preparations have become a revolutionary therapeutic approach that converts intestinal bacteria into cellular factories for medicinal chemical synthesis. The present paper discusses various aspects of engineering bacteria preparations, including wrapping materials, biomedical uses, and future developments.

3D magnetic flower-spherical Fe2O3-NiO derived from NiFe-layered double hydroxides for the efficient removal of Bensulfuron methyl: Morphological control and bimetallic synergy.

Three-dimensional (3D) magnetic flower-spherical Fe2O3-NiO derived from NiFe-layered double hydroxides (NiFe-LDHs) was fabricated through urea hydrothermal and calcination methods. The as-prepared materials were applied to activate PMS to degrade one of herbicide named Bensulfuron methyl (BSM). Fe2O3-NiO-1 demonstrated the highest catalytic activity and the lowest ions leaching by comparing the performance of LDHs and derivative bimetallic oxide synthesized by co-precipitation method, urea hydrothermal method and direct calcination method. Based on the results of SEM, BET and CV, the high catalytic activity of Fe2O3-NiO-1 originated from 3D morphology, lager specific area and pore size and faster electron transfer capability. The factors influencing the degradation performance were investigated and 0.1 g·L-1 Fe2O3-NiO could effectively activate PMS (1 mmol·L-1) to completely remove 10 mg·L-1 BSM within 30 min at pH 7.0. In Fe2O3-NiO/PMS system, OH, SO4- and 1O2 were produced and contributed to the BSM removal according to the results of EPR and quenching experiments. In order to expand its application range, Fe2O3-NiO/PMS system was used to degrade aniline (AN), sulfamethoxazole (SMZ), phenacetin (PNT), bisphenol A (BPA) and 2,4,6-triclofen (2,4,6-TCP) and the results showed the degradation efficiency could reach 90 % or more. Additionally, the application of catalysts in different actual water samples and the ability of reuse were tested. Based on the strategies of bimetallic synergy and morphology control, Fe-based bimetallic oxides with 3D morphology were developed in this study, which could effectively enhance the catalytic activity and inhibit the dissolution of metal ions, providing the design ideas for the construction of efficient catalysts and the removal of complex organic pollutants.

The impact of 25-hydroxyvitamin D and calcium on the risk of age-related macular degeneration: A Mendelian randomization study.

BACKGROUND: The relationships between 25-hydroxyvitamin D (25(OH)D) and calcium and age-related macular degeneration (AMD) are unclear. OBJECTIVE: This study aimed to investigate the causal role of 25(OH)D concentrations, calcium concentrations, and dietary supplements use of vitamin D and calcium on the risk of AMD and its subtypes. METHODS: Independent genetic variants associated with 25(OH)D and calcium concentrations were used as instrumental variables in published genome-wide association studies (GWASs) of European ancestry. The bidirectional two-sample Mendelian randomization (MR) analyses were performed using summary-level data from the UK Biobank and FinnGen datasets. Sensitivity analyses were conducted to ensure the robustness of the MR results. The meta-analyses were conducted using both fixed-effect and random-effect models to provide comprehensive and reliable estimates. RESULTS: A standard deviation increase in calcium concentrations was linked to a 14%, 17%, and 13% reduction in the likelihood of developing AMD (95% confidence interval [CI] = 0.77, 0.97), wet AMD (95% CI = 0.73, 0.95), and dry AMD (95% CI = 0.75, 1.00), respectively. No significant causal relationships were detected between genetically predicted 25(OH)D concentrations and AMD and its subtypes (all P > 0.05). The combined analyses showed that higher calcium concentrations were associated with a reduced risk of overall AMD, with an OR of 0.89 (95% CI = 0.81, 0.98). CONCLUSIONS: This study provides evidence supporting the causal relationship between calcium concentrations and the risk of AMD and its subtypes, which may have important implications for the prevention, monitoring, and treatment of AMD.

Monitoring landfill volatile organic compounds emissions by an uncrewed aerial vehicle platform with infrared and visible-light cameras, remote monitoring, and sampling systems.

An uncrewed aerial vehicle (UAV) platform equipped with dual imaging cameras, a gas sampling system, and a remote synchronous monitoring system was developed to sample and analyze volatile organic compounds (VOCs) emitted from landfills. The remote synchronous monitoring system provided real-time video to administrators with specific permissions to assist in identifying sampling sites within extensive landfill areas. The sampling system included four kits capable of collecting samples from different locations during a single flight mission. Each kit comprised a 1 L Tedlar bag for measuring landfill VOC concentrations according to the TO-15 method prescribed by the US Environmental Protection Agency. The air sample was introduced into a Tedlar bag via pumping. A known volume of the sample was subsequently concentrated using a solid multisorbent concentrator. Following this, the sample underwent cold trap concentration and thermal desorption. The concentrated sample was then transferred to a chromatography-mass spectrometry system for separation and analysis. Since the anaerobic catabolism of organic waste is exothermic and emits VOCs, this study employed UAV thermal imaging to locate principal emission sources for sampling. Visible-light imaging helped identify newer or older landfill sections, aiding in the selection of appropriate sampling sites, particularly when surfaces were thermally disturbed by solar radiation. Field measurements were conducted under three meteorological conditions: sunny morning, cirrus morning, and thin cloud evening (2 h after sunset), identifying 119, 122, and 111 chemical species respectively. The sequence of total VOC concentrations measured correlated with the meteorological conditions as follows: cirrus morning > thin cloud evening > sunny morning. The results indicated that ambient temperature and global solar radiation significantly influenced daytime gas emissions from landfills. Evening thermal images, unaffected by solar heating, facilitated more accurate identification of major VOC emission points, resulting in higher VOC concentrations compared to those recorded in the sunny morning. VOCs from the landfill were categorized into nine organic groups: alkanes, alkenes, carbonyls, aromatics, alcohols, esters, ethers, organic oxides, and others. The classification was based on carbon-containing compounds (Cn, where the compound contains n carbon atoms). Alkanes were predominant in terms of Cn presence, followed by alcohols and carbonyls. Among the organic groups, organic oxides, particularly 2-heptyl-1,3-dioxolane, exhibited the highest concentrations, succeeded by alkenes. Sampling under cloudy conditions or in the evening is recommended to minimize the effects of surface temperature anomalies caused by solar radiation, which vary due to differences in land composition.

Cell signaling and transcriptional regulation of osteoblast lineage commitment, differentiation, bone formation, and homeostasis.

The initiation of osteogenesis primarily occurs as mesenchymal stem cells undergo differentiation into osteoblasts. This differentiation process plays a crucial role in bone formation and homeostasis and is regulated by two intricate processes: cell signal transduction and transcriptional gene expression. Various essential cell signaling pathways, including Wnt, BMP, TGF-ß, Hedgehog, PTH, FGF, Ephrin, Notch, Hippo, and Piezo1/2, play a critical role in facilitating osteoblast differentiation, bone formation, and bone homeostasis. Key transcriptional factors in this differentiation process include Runx2, Cbfß, Runx1, Osterix, ATF4, SATB2, and TAZ/YAP. Furthermore, a diverse array of epigenetic factors also plays critical roles in osteoblast differentiation, bone formation, and homeostasis at the transcriptional level. This review provides an overview of the latest developments and current comprehension concerning the pathways of cell signaling, regulation of hormones, and transcriptional regulation of genes involved in the commitment and differentiation of osteoblast lineage, as well as in bone formation and maintenance of homeostasis. The paper also reviews epigenetic regulation of osteoblast differentiation via mechanisms, such as histone and DNA modifications. Additionally, we summarize the latest developments in osteoblast biology spurred by recent advancements in various modern technologies and bioinformatics. By synthesizing these insights into a comprehensive understanding of osteoblast differentiation, this review provides further clarification of the mechanisms underlying osteoblast lineage commitment, differentiation, and bone formation, and highlights potential new therapeutic applications for the treatment of bone diseases.

Novel progressive deep learning algorithm for uncovering multiple single nucleotide polymorphism interactions to predict paclitaxel clearance in patients with nonsmall cell lung cancer.

Background: The rate at which the anticancer drug paclitaxel is cleared from the body markedly impacts its dosage and chemotherapy effectiveness. Importantly, paclitaxel clearance varies among individuals, primarily because of genetic polymorphisms. This metabolic variability arises from a nonlinear process that is influenced by multiple single nucleotide polymorphisms (SNPs). Conventional bioinformatics methods struggle to accurately analyze this complex process and, currently, there is no established efficient algorithm for investigating SNP interactions. Methods: We developed a novel machine-learning approach called GEP-CSIs data mining algorithm. This algorithm, an advanced version of GEP, uses linear algebra computations to handle discrete variables. The GEP-CSI algorithm calculates a fitness function score based on paclitaxel clearance data and genetic polymorphisms in patients with nonsmall cell lung cancer. The data were divided into a primary set and a validation set for the analysis. Results: We identified and validated 1184 three-SNP combinations that had the highest fitness function values. Notably, SERPINA1, ATF3 and EGF were found to indirectly influence paclitaxel clearance by coordinating the activity of genes previously reported to be significant in paclitaxel clearance. Particularly intriguing was the discovery of a combination of three SNPs in genes FLT1, EGF and MUC16. These SNPs-related proteins were confirmed to interact with each other in the protein-protein interaction network, which formed the basis for further exploration of their functional roles and mechanisms. Conclusion: We successfully developed an effective deep-learning algorithm tailored for the nuanced mining of SNP interactions, leveraging data on paclitaxel clearance and individual genetic polymorphisms.

Deep Eutectic Solvent-Assisted Corrosion Boosting Bulk FeCoNiCrMo High-Entropy Alloys as Highly Efficient Oxygen Evolution Reaction Catalyst.

The key to enhancing water electrolysis efficiency lies in selecting highly efficient catalysts. Currently, high-entropy alloys (HEAs) are utilized in electrocatalysis applications owing to their diverse elemental composition, disordered elemental distribution, and the high solubility of each element, endowing them with excellent catalytic performance. The experiments were conducted using isoatomic FeNiCrMo HEA as a precursor, with a high-activity three-dimensional nanoporous structure rapidly synthesized via electrochemical one-step dealloying in a choline chloride-thiourea (ChCl-TU) deep eutectic solvent (DES). The results indicate that the dealloyed Fe20Co20Ni20Cr20Mo20 HEA mainly consists of two phases: face-centered cubic and σ phases. The imbalance in the distribution of elements in these two phases leads to quite different corrosion speeds with the FCC phase being preferentially corroded. Furthermore, synergistic electron coupling between surface atoms in the three-dimensional nanoporous structure strengthens the behavior of the oxygen evolution reaction (OER). At a current density of 40 mA cm-2, the overpotential after dealloying decreased to 370 mV, demonstrating excellent stability. The technique demonstrated in this work provides a novel approach to improve the catalytic activity of OER.

Electro-optically tunable optical delay on a lithium niobate photonic chip.

An approach for continuous tuning of on-chip optical delay with a microring resonator is proposed and demonstrated. By introducing an electro-optically tunable waveguide coupler, the bus waveguide to the resonance coupling can be effectively tuned from the under-coupling regime to the over-coupling regime. The optical delay is experimentally characterized by measuring the relative phase shift between lasers and shows a large dynamic range of delay from -600 to 600 ps and an efficient tuning of delay from -430 to -180 ps and from 40 to 240 ps by only a 5 V voltage.

The dual nature of working memory deficits: methamphetamine abusers have more impaired social working memory capacity than canonical working memory capacity.

Social working memory (WM) temporarily retains and manipulates various aspects of social information. Extensive research has highlighted impaired social cognitive functions in individuals with substance addiction. However, the specific deficit in social WM within this population remains notably understudied. Bridging this gap, we investigated social WM capacity using biological motion (BM) stimuli in methamphetamine (MA) abusers compared to an inmate control group, alongside contrasting these findings with their canonical WM deficits. Across two studies, we recruited female MA abusers (N = 80) undergoing post-isolation rehabilitation within a mandatory confinement circumstance. To ensure a pertinent comparison, we recruited female inmates (N = 80) subjected to comparable confinement. Results show substantial BM WM impairment in MA abusers, yet non-BM WM remains mostly intact. These findings highlight a pronounced social WM deficit in MA abusers, surpassing their canonical WM deficit relative to inmate controls. This suggests a distinct dissociation between social and canonical WM processing.

Activin A as a potential biomarker for preserved ratio impaired spirometry (PRISm) and clinical outcomes in community-dwelling adults.

INTRODUCTION: This study endeavors to decipher the association between Activin A and PRISm, thereby addressing the potential of Activin A as a serum biomarker for early detection and long-term clinical outcome prediction of PRISm and subsequent all-cause mortality. METHODS: The study sample comprised middle-aged and older adults from the I-Lan Longitudinal Aging Study. Pulmonary function including forced vital capacity (FVC) and forced expiratory volume in one second (FEV1) were measured. Demographic data and laboratory data (including serum Activin A levels) were also collected. Multivariate logistic regression and Cox proportional hazards models were used to identify independent predictors of PRISm and all-cause mortality, respectively. RESULTS: Among 711 eligible participants, 34 % had PRISm. The risk of PRISm elevated with Activin A levels in group quartiles (adjusted odds ratio (aOR), Q2: 1.606 [95 % CI 0.972-2.652], p = 0.064, Q3: 2.666 [1.635-4.348], p < 0.001, Q4: 3.225 [1.965-5.293], p < 0.001). On the other hand, lower hemoglobin (aOR: 1.122, p = 0.041) and higher blood urea nitrogen (BUN) levels (aOR: 1.033, p = 0.048) were associated with increased risk of PRISm. In addition, the PRISm group had a higher all-cause mortality rate (non-PRISm 4.5% vs. PRISm 8.3 %, p = 0.038). Multivariate Cox models also identify a higher level of Activin A as a risk factor of all-cause mortality (aHR: 1.001 [1.000-1.003], p = 0.042). CONCLUSIONS: Higher Activin A quartiles were linked to increased risk of PRISm, along with lower hemoglobin and higher BUN levels. Additonally, elevated Activin A was a significant risk factor of all-cause mortality.

Irisin attenuates acute glaucoma-induced neuroinflammation by activating microglia-integrin αVß5/AMPK and promoting autophagy.

Neuroinflammation, characterized by microglial activation and the release of multiple inflammatory mediators, is a key factor in acute glaucomatous injury leading to retinal ganglion cell (RGC) death and ultimately irreversible vision loss. Irisin, a novel exercise-induced myokine, has demonstrated anti-inflammatory activity in ischemia/reperfusion injuries across multiple organs and has displayed a significant neuroprotective role in experimental stroke disease models. This study examined the protective impact of irisin and investigated its potential mechanism involved in this process utilizing an acute ocular hypertension (AOH)-induced retinal injury model in mice and a microglia inflammation model induced by lipopolysaccharide (LPS). There was a transient downregulation of irisin in the retina after AOH injury, with parallel emergence of retinal neuroinflammation and RGC death. Irisin attenuated retinal and optic nerve damage and promotes the phenotypic conversion of microglia from M1 to M2. Mechanistically, irisin significantly upregulated the expression of integrin αVß5, p-AMPK, and autophagy-related markers. Integrin αVß5 was highly expressed on microglia but hardly expressed on RGC. The integrin αVß5 inhibitor cilengitide, the AMPK inhibitor dorsomorphin, and the autophagy inhibitor 3-Methyladenine (3-MA) blocked the neuroprotective effects of irisin. Our results suggest irisin attenuates acute glaucoma-induced neuroinflammation and RGC death by activating integrin αVß5/AMPK in microglia and promoting autophagy. It should be considered a potential neuroprotective therapy for acute glaucoma.

DeepSAP: A Novel Brain Image-Based Deep Learning Model for Predicting Stroke-Associated Pneumonia From Spontaneous Intracerebral Hemorrhage.

RATIONALE AND OBJECTIVE: Stroke-associated pneumonia (SAP) often appears as a complication following intracerebral hemorrhage (ICH), leading to poor prognosis and increased mortality rates. Previous studies have typically developed prediction models based on clinical data alone, without considering that ICH patients often undergo CT scans immediately upon admission. As a result, these models are subjective and lack real-time applicability, with low accuracy that does not meet clinical needs. Therefore, there is an urgent need for a quick and reliable model to timely predict SAP. METHODS: In this retrospective study, we developed an image-based model (DeepSAP) using brain CT scans from 244 ICH patients to classify the presence and severity of SAP. First, DeepSAP employs MRI-template-based image registration technology to eliminate structural differences between samples, achieving statistical quantification and spatial standardization of cerebral hemorrhage. Subsequently, the processed images and filtered clinical data were simultaneously input into a deep-learning neural network for training and analysis. The model was tested on a test set to evaluate diagnostic performance, including accuracy, specificity, and sensitivity. RESULTS: Brain CT scans from 244 ICH patients (mean age, 60.24; 66 female) were divided into a training set (n = 170) and a test set (n = 74). The cohort included 143 SAP patients, accounting for 58.6% of the total, with 66 cases classified as moderate or above, representing 27% of the total. Experimental results showed an AUC of 0.93, an accuracy of 0.84, a sensitivity of 0.79, and a precision of 0.95 for classifying the presence of SAP. In comparison, the model relying solely on clinical data showed an AUC of only 0.76, while the radiomics method had an AUC of 0.74. Additionally, DeepSAP achieved an optimal AUC of 0.84 for the SAP grading task. CONCLUSION: DeepSAP's accuracy in predicting SAP stems from its spatial normalization and statistical quantification of the ICH region. DeepSAP is expected to be an effective tool for predicting and grading SAP in clinic.