Olanzapine vs. magnesium valproate vs. lamotrigine in anti-N-methyl-D-aspartic acid receptor encephalitis: a retrospective study.

BACKGROUND: This study aimed to compare the impact of olanzapine, magnesium valproate, and lamotrigine as adjunctive treatments for anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis. And it is expected to add supporting points related to the rebalance of neurotransmitters in the brain through adjuvant therapy in the clinical management of anti-NMDAR encephalitis. METHODS: This retrospective study included patients diagnosed with anti-NMDAR encephalitis who received standardized immunotherapy at Hunan Brain Hospital between January 2018 and December 2020. RESULTS: Compared to the olanzapine group, both the magnesium valproate and lamotrigine groups showed lower scores on the positive and negative symptom scale (PANSS) total score after 3 weeks of treatment (all P < 0.05). The Montreal Cognitive Assessment Scale (MoCA) scores in the magnesium valproate and lamotrigine groups were significantly higher than in the olanzapine group after 3 weeks and 3 months of treatment (all P < 0.05). After 3 months of treatment, the proportions of patients with a modified Rankin scale score (mRS) of 0-1 in the magnesium valproate and lamotrigine groups were significantly higher than in the olanzapine group (all P < 0.05). The electroencephalogram (EEG) abnormality ranks at 3 months were significantly lower in the magnesium valproate and lamotrigine groups compared with the olanzapine group (all P < 0.05). Furthermore, the Glx/Cr ratio significantly decreased after 3 months of treatment (all P < 0.05) in the magnesium valproate and lamotrigine groups, while the Glx/Cr ratio in the olanzapine group showed no significant change (P > 0.05). CONCLUSION: Compared with olanzapine, the addition of magnesium valproate or lamotrigine to immunotherapy might be associated with a lower PANSS score, higher MoCA score, and lower mRS score. The improvement of neurological functions and cognitive function may be related to the decreased Glx/Cr ratio.

A Comprehensive Proteogenomic and Spatial Analysis of Innate and Acquired Resistance of Metastatic Melanoma to Immune Checkpoint Blockade Therapies.

While a subset of patients with metastatic melanoma achieves durable responses to immune checkpoint blockade (ICB) therapies, the majority ultimately exhibit either innate or acquired resistance to these treatments. However, the molecular mechanisms underlying resistance to ICB therapies remain elusive and are warranted to elucidate. Here, we comprehensively investigated the tumor and tumor immune microenvironment (TIME) of paired pre- and post-treatment tumor specimens from metastatic melanoma patients who were primary or secondary resistance to anti-CTLA-4 and/or anti-PD-1/PD-L1 therapies. Differentially expressed gene (DEG) analysis and single-sample gene set enrichment analysis (ssGSEA) with transcriptomic data identified cell cycle and c-MYC signaling as pathway-based resistance signatures. And weighted gene co-expression network analysis (WGCNA) revealed the activation of a cross-resistance meta-program involving key signaling pathways related to tumor progression in ICB resistant melanoma. Moreover, spatially-resolved, image-based immune monitoring analysis by using NanoString's digital spatial profiling (DSP) and Cyclic Immunofluorescence (CyCIF) showed infiltration of suppressive immune cells in the tumor microenvironment of melanoma with resistance to ICB therapies. Our study reveals the molecular mechanisms underlying resistance to ICB therapies in patients with metastatic melanoma by conducting such integrated analyses of multi-dimensional data, and provides rationale for salvage therapies that will potentially overcome resistance to ICB therapies. Statement of translational relevance: This study paves the way for the creation of innovative therapeutic strategies, aimed at subverting resistance to immune checkpoint blockade (ICB) therapies in metastatic melanoma patients. By unraveling the specific molecular mechanisms underlying resistance, scientists can design effective alternative treatments that target pathways such as pathways associated with cell cycle dysregulation and c-MYC signaling. Furthermore, through the application of advanced immune monitoring techniques such as NanoString Digital Spatial Profiling (DSP) and Cyclic Immunofluorescence (CyCIF), this study has significantly enriched our understanding of the tumor microenvironment. This enhanced characterization facilitates the discovery of potential biomarkers that may forecast a patient's response to ICB treatment. Ultimately, these advancements could potentially refine patient outcomes and foster the development of more personalized cancer treatments in the future.

Identification of Potential Therapeutics for Infantile Hemangioma via in silico Investigation and in vitro Validation.

Introduction: Infantile Hemangioma (IH) is a prevalent benign vascular tumor affecting approximately 5-10% of infants. Its underlying pathogenesis remains enigmatic, and current therapeutic approaches show limited effectiveness. Our study aimed to discover potential IH-associated therapeutics through a transcriptomic, computational drug repurposing methodology. Methods: Utilizing the IH-specific dataset GSE127487 from the Gene Expression Omnibus, we identified differentially expressed genes (DEGs) and conducted weighted gene coexpression network analysis (WGCNA). Subsequently, a protein-protein interaction (PPI) network was constructed to obtain the top 100 hub genes. Drug candidates were sourced from the Connectivity Map (CMap) and Comparative Toxicogenomics Database (CTD). Results: Our analysis revealed 1203 DEGs and a significant module of 1780 mRNAs strongly correlated with IH. These genes were primarily enriched in the PI3K/AKT/MTOR, RAS/MAPK, and CGMP/PKG signaling pathway. After creating a PPI network of overlapping genes, we filtered out the top 100 hub genes. Ultimately, 44 non-toxic drugs were identified through the CMap and CTD databases. Twelve molecular-targeting agents (belinostat, chir 99021, dasatinib, entinostat, panobinostat, sirolimus, sorafenib, sunitinib, thalidomide, U 0126, vorinostat, and wortmannin) may be potential candidates for IH therapy. Moreover, in vitro experiments demonstrated that entinostat, sorafenib, dasatinib, and sirolimus restricted the proliferation and migration and initiated apoptosis in HemEC cells, thereby underscoring their potential therapeutic value. Conclusion: Our investigation revealed that the pathogenic mechanism underlying IH might be closely associated with the PI3K/AKT/MTOR, RAS/MAPK, and CGMP/PKG signaling pathways. Furthermore, we identified twelve molecular-targeting agents among the predicted drugs that show promise as therapeutic candidates for IH.

Transcriptomic analysis used to discover potential therapeutics for Infantile Hemangioma (IH). Key IH-related pathways: PI3K/AKT/MTOR, RAS/MAPK, and CGMP/PKG signaling identified. Identified 44 non-toxic drugs as potential IH therapies via CMap and CTD. Twelve molecular agents show potential as IH therapy candidates. In vitro studies confirmed entinostat, sorafenib, dasatinib, and sirolimus inhibit HemEC cell proliferation and induce apoptosis.

A Novel Multi-Dimensional Joint Search Method for the Compression of Medical Image Segmentation Models.

Due to the excellent results achieved by transformers in computer vision, more and more scholars have introduced transformers into the field of medical image segmentation. However, the use of transformers will make the model's parameters very large, which occupies a large amount of the computer's resources, making them very time-consuming during training. In order to alleviate this disadvantage, this paper explores a flexible and efficient search strategy that can find the best subnet from a continuous transformer network. The method is based on a learnable and uniform L1 sparsity constraint, which contains factors that reflect the global importance of the continuous search space in different dimensions, while the search process is simple and efficient, containing a single round of training. At the same time, in order to compensate for the loss of accuracy caused by the search, a pixel classification module is introduced into the model to compensate for the loss of accuracy in the model search process. Our experiments show that the model in this paper compresses 30% of the parameters and FLOPs used, while also showing a slight increase in the accuracy of the model on the Automatic Cardiac Diagnosis Challenge (ACDC) dataset.

The destruction of cytoplasmic skeleton leads to the change of nuclear structure and the looseness of lamin A submicroscopic network.

The interaction between lamin A and the cytoplasmic skeleton plays a key role in maintaining nuclear mechanical properties. However, the effect of destruction of the cytoplasmic skeleton on the 3D submicroscopic structure of lamin A has not been elucidated. In this study, we developed an image quantization algorithm to quantify changes in the submicroscopic structure of the intact lamin A 3D network within the nucleus. We used blebbistatin or nocodazole to disrupt the fibrillar structure of F-actin or tubulin, respectively, and then quantified changes in the lamin A super-resolution network structure, the morphological and mechanical properties of the nucleus and the spatial distribution of chromosomes. Ultimately, we found for the first time that disruption of the cytoplasmic skeleton changes the lamin A submicroscopic network and nuclear structural characteristics. In summary, this study contributes to understanding the trans-nuclear membrane interaction characteristics of lamin A and the cytoplasmic skeleton.

Association between breastfeeding duration and neurodevelopment in Chinese children aged 2 to 3 years.

OBJECTIVE: The aim of this study was to explore the relationship between breastfeeding duration and neurodevelopment in children aged 2 to 3 years in a Chinese population. METHODS: This study was based on a cross-sectional survey. The data were from the National Nutrition and Health Systematic Survey for children in China which was conducted from 2019 to 2020. Characteristics of parents and children and the breastfeeding duration were obtained using interview-administered questionnaires. Children's neuropsychological development was assessed by a trained child health care physician using the Child Psychological Development Scale. A multivariable linear regression model was used to analyze the association between breastfeeding duration and neuropsychological development. RESULTS: A total 1290 children aged 2-3 years were included in the present analysis. In multivariable linear regression models, after adjustment for potential confounders, children who were breastfed for 7-12 months had a 3.59-point increase in gross motor development (ß = 3.59; 95 % confidence interval [CI]: 1.23 to 6.34), a 3.73-point increase in fine motor development (ß = 3.73; 95 % CI: 1.09 to 6.47), and a 2.87-point in language development (ß = 2.87; 95 % CI: 1.12 to 5.31) compared with those who were never breastfed. Children who were breastfed for > 12 months had a 3.77-point increase in fine motor development (ß = 3.77; 95 % CI: 0.98 to 6.86) compared with those who were never breastfed. CONCLUSIONS: Longer breastfeeding duration was associated with increased gross motor, fine motor, and language scores in our study population. Mothers in China should be encouraged to initiate and continue breastfeeding.

A novel 3D-printed educational model for the training of laparoscopic bile duct Exploration:a pilot study for beginning trainees.

Background: Laparoscopic common bile duct exploration (LCBDE) is a minimally invasive procedure for the removal of bile duct stones that is often performed by experienced hepatobiliary surgeons; beginners do not easily master this approach. Aim: To investigate the effectiveness and practicality of a three-dimensional printed (3DP) anatomical model based on radiographic images in the training of LCBDE techniques and formulate standardized educational workflows. Methods: Colored LCBDE training models were produced using 3DP technology. Twenty standardized training trainees were randomly divided into two groups: a 3DP model training group and a traditional laparoscopic simulation training group. Both groups received the same number of teaching hours. After a 4-weeks training course, the trainees' subjective and objective progress in basic knowledge and manipulations were evaluated and compared. Results: Compared with traditional laparoscopic simulation, 3DP model simulation training is of great significance in improving trainers' understanding of surgical procedures and cooperation during the operation. Trainees with 3DP models training demonstrated a significant improvement in their understanding of the key points of surgery (χ2 = 6.139, p = 0.013) and skills scores, especially in core procedural steps operation. More importantly, the trainees showed higher levels of satisfaction and self-confidence while assisting in the surgery. Conclusion: With the development of 3DP models, improvements in the learning effect for theoretical understanding and practical skills were significant. The present study is the initial educational experience with 3DP models to facilitate the operational capabilities of the trainees for LCBDE.

From molecular subgroups to molecular targeted therapy in rheumatoid arthritis: A bioinformatics approach.

1Background: Rheumatoid Arthritis (RA) is a heterogeneous autoimmune disease with multiple unidentified pathogenic factors. The inconsistency between molecular subgroups poses challenges for early diagnosis and personalized treatment strategies. In this study, we aimed to accurately distinguish RA patients at the transcriptome level using bioinformatics methods. 2Methods: We collected a total of 362 transcriptome datasets from RA patients in three independent samples from the GEO database. Consensus clustering was performed to identify molecular subgroups, and clinical features were assessed. Differential analysis was employed to annotate the biological functions of specifically upregulated genes between subgroups. 3Results: Based on consensus clustering of RA samples, we identified three robust molecular subgroups, with Subgroup III representing the high-risk subgroup and Subgroup II exhibiting a milder phenotype, possibly associated with relatively higher levels of autophagic ability. Subgroup I showed biological functions mainly related to viral infections, cellular metabolism, protein synthesis, and inflammatory responses. Subgroup II involved autophagy of mitochondria and organelles, protein localization, and organelle disassembly pathways, suggesting heterogeneity in the autophagy process of mitochondria that may play a protective role in inflammatory diseases. Subgroup III represented a high-risk subgroup with pathological processes including abnormal amyloid precursor protein activation, promotion of inflammatory response, and cell proliferation. 4Conclusion: The classification of the RA dataset revealed pathological heterogeneity among different subgroups, providing new insights and a basis for understanding the molecular mechanisms of RA, identifying potential therapeutic targets, and developing personalized treatment approaches.

Flexible Crossbar Molecular Devices with Patterned EGaIn Top Electrodes for Integrated All-Molecule-Circuit Implementation.

Here, a unique crossbar architecture is designed and fabricated, incorporating vertically integrated self-assembled monolayers in electronic devices. This architecture is used to showcase 100 individual vertical molecular junctions on a single chip with a high yield of working junctions and high device uniformity. The study introduces a transfer approach for patterned liquid-metal eutectic alloy of gallium and indium top electrodes, enabling the creation of fully flexible molecular devices with electrical functionalities. The devices exhibit excellent charge transport performance, sustain a high rectification ratio (>103), and stable endurance and retention properties, even when the devices are significantly bent. Furthermore, Boolean logic gates, including OR and AND gates, as well as half-wave and full-wave rectifying circuits, are successfully implemented. The unique design of the flexible molecular device represents a significant step in harnessing the potential of molecular devices for high-density integration and possible molecule-based computing.

The association between macrosomia and glucose, lipids and hormones levels in maternal and cord serum: a case-control study.

BACKGROUND: The formation of macrosomia is associated with excessive nutrition and/or unable to regulate effectively. This case-control study aims to explore the relationship between macrosomia and glucose, lipids and hormones levels in maternal and cord serum. METHODS: In the case-control study, 78 pairs of mothers and newborns were recruited who received care at one hospital of Hebei, China between 2016 and 2019. According to the birth weight (BW) of newborns, participants were divided into macrosomia group (BW ≥ 4000 g, n = 39) and control group (BW between 2500 g and 3999 g, n = 39). Maternal vein blood and cord vein blood were collected and assayed. All data were compared between the two groups. Unconditional logistics regression analysis was used to test the relationship between macrosomia and glucose, lipids and hormones in maternal and cord serum. RESULTS: In maternal and cord serum, the levels of leptin, leptin/adiponectin ratio (LAR), glucose and triglyceride (TG) in macrosomia group were higher than those in control group, and the levels of high-density lipoprotein cholesterol (HDL-C) were lower. The percentage of maternal glucose and lipids transfer to cord blood did not differ between the two groups. High levels of TG in maternal serum were positively correlated with macrosomia, and high levels of LAR, TG and glucose in cord serum were positively correlated with macrosomia. CONCLUSION: In conclusion, the results of the current study, suggest that the nutrients and metabolism-related hormones in maternal and umbilical cord are closely related to macrosomia. During pregnancy, the nutritional status of pregnant women should be paid attention to and to obtain a good birth outcome.

Different care mode alter composition and function of gut microbiota in cerebral palsy children.

Introduction: Specialized care is essential for the recovery of children with cerebral palsy (CP). This study investigates how different care modes impact the gut microbiota. Methods: Fecal samples from 32 children were collected, among whom those cared for by family (n = 21) were selected as the observation group, and those cared for by children's welfare institutions (n = 11) were selected as the control group (registration number of LGFYYXLL-024). The gut microbiota profiles were analyzed. Results: There was no significant difference in the α-diversity of the gut microbiota and the abundance at the phylum level. However, at the genus level, the observation group showed a significant increase in the abundance of butyrate-producing bacteria Bacteroides and Lachnospiracea incertae sedis (P < 0.05), and a significant decrease in the abundance of opportunistic pathogens Prevotella, Clostridium cluster IV, Oscillibacter, and Fusobacterium (P < 0.05). Additionally, lipid metabolism, carbohydrate metabolism, transcription, cellular processes and signaling, and membrane transport were significantly upregulated in the observation group. Lipid metabolism was positively correlated with Bacteroides and Lachnospiracea incertae sedis, indicating a positive impact of the family-centered care mode on bacterial metabolism processes. Discussion: This study highlights that the family-centered care mode had a positive impact on the composition and function of the gut microbiota. The study provides valuable insights into the relationship between care mode and gut microbiota, which can inspire the development of interventions for cerebral palsy.

A quinine-squaramide catalyzed enantioselective vinylogous Mannich reaction between benzothiazolimines and γ-butenolides for efficient preparation of chiral N-benzothiazole butyrolactones.

A highly effective and enantioselective vinylogous Mannich reaction between benzothiazolimines and γ-butenolides catalyzed by a quinine based squaramide has been disclosed. A series of chiral benzothiazole amines containing a γ,γ-disubstituted butanolide scaffold bearing an adjacent chiral stereocenter have been successfully obtained in good to excellent yields (up to 91%) with excellent enantioselectivities (up to >99% ee) and diastereoselectivities (>20 : 1 dr) with broad substrate generality under mild conditions. The new scaffold integrated with both chiral benzothiazolimine and γ-butenolide moieties may provide a possibility for the development of new pharmaceutical entities.

[Research progress of traditional Chinese medicine in regulating autophagy for intervening in diabetic cardiomyopathy].

Diabetic cardiomyopathy(DCM) is a chronic complication of diabetes mellitus that leads to cardiac damage in the later stages of the disease, and its pathogenesis is complex, involving metabolic disorders brought about by a variety of aberrant alterations such as endoplasmic reticulum stress, oxidative stress, inflammation, and apoptosis, defects in cardiomyocyte Ca~(2+) transporter, and myocardial fibrosis. Currently, there is a lack of specific diagnosis and treatment in the clinic. Autophagy is a highly conserved scavenging mechanism that removes proteins, damaged organelles or foreign contaminants and converts them into energy and amino acids to maintain the stability of the intracellular environment. Inhibition of autophagy can cause harmful metabolites to accumulate in the cell, while over-activation of autophagy can disrupt normal cellular structures and cause cell death. Prolonged high glucose levels disrupt cardiomyocyte autophagy levels and exacerbate the development of DCM. The protective or detrimental effects of autophagy on cells ring true with the traditional Chinese medicine theory of healthy Qi and pathogenic Qi. Autophagy in the physiological state of the removal of intracellular substances and the generation of substances beneficial to the survival of cells is the inhibition of pathogenic Qi to help the performance of healthy Qi, so the organism is healthy. In the early stages of the disease, when autophagy is impaired and incapable of removing waste substances, pathogenic Qi is prevalent; In the later stages of the disease, excessive activation of autophagy can destroy normal cells, leading to a weakening of healthy Qi. Traditional Chinese medicine has the advantage of targeting multiple sites and pathways. Studies in recent years have confirmed that traditional Chinese medicine monomers or formulas can target autophagy, promote the restoration of autophagy levels, maintain mitochondrial and endoplasmic reticulum homeostasis, and reduce oxidative stress, endoplasmic reticulum stress, inflammation, and apoptosis in order to prevent and control DCM. This study provides a review of the relationship between autophagy and DCM and the intervention of traditional Chinese medicine in autophagy for the treatment of DCM, with a view to providing new clinical ideas and methods for the treatment of DCM with traditional Chinese medicine.

Dynamically blocking leakage current in molecular tunneling junctions.

Molecular tunneling junctions based on self-assembled monolayers (SAMs) have demonstrated rectifying characteristics at the nanoscale that can hardly be achieved using traditional approaches. However, defects in SAMs result in high leakage when applying bias. The poor performance of molecular diodes compared to silicon or thin-film devices limits their further development. In this study, we show that incorporating "mixed backbones" with flexible-rigid structures into molecular junctions can dynamically block tunneling currents, which is difficult to realize using non-molecular technology. Our idea is achieved by the interaction between interfacial dipole moments and electric field, triggering structured packing in SAMs. Efficient blocking of leakage by more than an order of magnitude leads to a significant enhancement of the rectification ratio to the initial value. The rearrangement of supramolecular structures has also been verified through electrochemistry and electroluminescence measurements. Moreover, the enhanced rectification is extended to various challenging environments, including endurance measurements, bending of electrodes, and rough electrodes, thus demonstrating the feasibility of the dynamic behavior of molecules for practical electronic applications.

Low-light enhancement method with dual branch feature fusion and learnable regularized attention.

Restricted by the lighting conditions, the images captured at night tend to suffer from color aberration, noise, and other unfavorable factors, making it difficult for subsequent vision-based applications. To solve this problem, we propose a two-stage size-controllable low-light enhancement method, named Dual Fusion Enhancement Net (DFEN). The whole algorithm is built on a double U-Net structure, implementing brightness adjustment and detail revision respectively. A dual branch feature fusion module is adopted to enhance its ability of feature extraction and aggregation. We also design a learnable regularized attention module to balance the enhancement effect on different regions. Besides, we introduce a cosine training strategy to smooth the transition of the training target from the brightness adjustment stage to the detail revision stage during the training process. The proposed DFEN is tested on several low-light datasets, and the experimental results demonstrate that the algorithm achieves superior enhancement results with the similar parameters. It is worth noting that the lightest DFEN model reaches 11 FPS for image size of 1224×1024 in an RTX 3090 GPU.

Preferentially expressed endosperm genes reveal unique activities in wheat endosperm during grain filling.

BACKGROUND: Bread wheat (Triticum aestivum L.) endosperm contains starch and proteins, which determine the final yield, quality, and nutritional value of wheat grain. The preferentially expressed endosperm genes can precisely provide targets in the endosperm for improving wheat grain quality and nutrition using modern bioengineering technologies. However, the genes specifically expressed in developing endosperms remain largely unknown. RESULTS: In this study, 315 preferentially expressed endosperm genes (PEEGs) in the spring wheat landrace, Chinese Spring, were screened using data obtained from an open bioinformatics database, which reveals a unique grain reserve deposition process and special signal transduction in a developing wheat endosperm. Furthermore, transcription and accumulation of storage proteins in the wheat cultivar, XC26 were evaluated. The results revealed that 315 PEEG plays a critical role in storage protein fragment deposition and is a potential candidate for modifying grain quality and nutrition. CONCLUSION: These results provide new insights into endosperm development and candidate genes and promoters for improving wheat grain quality through genetic engineering and plant breeding techniques.

Epidemiological Characteristics of Fracture Among Children Aged 6-17 Years - China, 2019-2021.

What is already known about this topic?: Fractures are a common and serious injury among children. While many studies have utilized clinical data, there is a lack of large-scale community-based research in China. What is added by this report?: This cross-sectional study provides national and regionally representative data on the prevalence of fractures among Chinese children aged 6-17 years (6.93%), with higher rates observed in males than in females (8.13% vs. 5.71%) and in rural areas compared to urban areas (7.22% vs. 6.62%). The most common site of fracture was the upper limbs (4.24%, accounting for 63.0% of fractures). What are the implications for public health practice?: The need to enhance awareness of fracture prevention is critical, particularly for children in rural areas and males in middle childhood. We recommend that local authorities increase investments in educational programs and child safety oversight. Additionally, promoting balanced diets for children, training in proper exercise techniques, and reinforcing participation in outdoor sports are essential.

Restoration of motion-blurred numeral image using a complex-amplitude diffractive processor.

We propose a complex-amplitude diffractive processor based on diffractive deep neural networks (D2NNs). By precisely controlling the propagation of an optical field, it can effectively remove the motion blur in numeral images and realize the restoration. Comparative analysis of phase-only, amplitude-only, and complex-amplitude diffractive processor reveals that the complex-amplitude network significantly enhances the performance of the processor and improves the peak signal-to-noise ratio (PSNR) of the images. Appropriate use of complex-amplitude networks contributes to reduce the number of network layers and alleviates alignment difficulties. Due to its fast processing speed and low power consumption, complex-amplitude diffractive processors hold potential applications in various fields including road monitoring, sports photography, satellite imaging, and medical diagnostics.

A dual-radiomics model for overall survival prediction in early-stage NSCLC patient using pre-treatment CT images.

Introduction: Radiation therapy (RT) is one of the primary treatment options for early-stage non-small cell lung cancer (ES-NSCLC). Therefore, accurately predicting the overall survival (OS) rate following radiotherapy is crucial for implementing personalized treatment strategies. This work aims to develop a dual-radiomics (DR) model to (1) predict 3-year OS in ES-NSCLC patients receiving RT using pre-treatment CT images, and (2) provide explanations between feature importanceand model prediction performance. Methods: The publicly available TCIA Lung1 dataset with 132 ES-NSCLC patients received RT were studied: 89/43 patients in the under/over 3-year OS group. For each patient, two types of radiomic features were examined: 56 handcrafted radiomic features (HRFs) extracted within gross tumor volume, and 512 image deep features (IDFs) extracted using a pre-trained U-Net encoder. They were combined as inputs to an explainable boosting machine (EBM) model for OS prediction. The EBM's mean absolute scores for HRFs and IDFs were used as feature importance explanations. To evaluate identified feature importance, the DR model was compared with EBM using either (1) key or (2) non-key feature type only. Comparison studies with other models, including supporting vector machine (SVM) and random forest (RF), were also included. The performance was evaluated by the area under the receiver operating characteristic curve (AUCROC), accuracy, sensitivity, and specificity with a 100-fold Monte Carlo cross-validation. Results: The DR model showed highestperformance in predicting 3-year OS (AUCROC=0.81 ± 0.04), and EBM scores suggested that IDFs showed significantly greater importance (normalized mean score=0.0019) than HRFs (score=0.0008). The comparison studies showed that EBM with key feature type (IDFs-only demonstrated comparable AUCROC results (0.81 ± 0.04), while EBM with non-key feature type (HRFs-only) showed limited AUCROC (0.64 ± 0.10). The results suggested that feature importance score identified by EBM is highly correlated with OS prediction performance. Both SVM and RF models were unable to explain key feature type while showing limited overall AUCROC=0.66 ± 0.07 and 0.77 ± 0.06, respectively. Accuracy, sensitivity, and specificity showed a similar trend. Discussion: In conclusion, a DR model was successfully developed to predict ES-NSCLC OS based on pre-treatment CT images. The results suggested that the feature importance from DR model is highly correlated to the model prediction power.