Circulating tumor cells with increasing aneuploidy predict inferior prognosis and therapeutic resistance in small cell lung cancer.

AIMS: Treatment resistance commonly emerges in small cell lung cancer (SCLC), necessitating the development of novel and effective biomarkers to dynamically assess therapeutic efficacy. This study aims to evaluate the clinical utility of aneuploid circulating tumor cells (CTCs) for risk stratification and treatment response monitoring. METHODS: A total of 126 SCLC patients (two cohorts) from two independent cancer centers were recruited as the study subjects. Blood samples were collected from these patients and aneuploid CTCs were detected. Aneuploid CTC count (ACC) and aneuploid CTC score (ACS), were used to predict progression-free survival (PFS) and overall survival (OS). The performance of the ACC and the ACS was evaluated by calculating the area under the receiver operating characteristic (ROC) curve (AUC). RESULTS: Compared to ACC, ACS exhibited superior predictive power for PFS and OS in these 126 patients. Moreover, both univariate and multivariate analyses revealed that ACS was an independent prognostic factor. Dynamic ACS changes reflected treatment response, which is more precise than ACC changes. ACS can be used to assess chemotherapy resistance and is more sensitive than radiological examination (with a median lead time of 2.8 months; P < 0.001). When patients had high ACS levels (> 1.115) at baseline, the combination of immunotherapy and chemotherapy resulted in longer PFS (median PFS, 7.7 months; P = 0.007) and OS (median OS, 16.3 months; P = 0.033) than chemotherapy alone (median PFS, 4.9 months; median OS, 13.6 months). CONCLUSIONS: ACS could be used as a biomarker for risk stratification, treatment response monitoring, and individualized therapeutic intervention in SCLC patients.

From the pitch to personal growth: Investigating self-esteem as a mediator and parental support as a moderator in youth sports in China.

This study focuses on the intricate connections among youth sports, personal development, and skill acquisition in contemporary China. Building upon established theoretical frameworks, the study aims to unravel the intricate interplay among various variables. Employing a robust methodology that accounts for mediation and moderation effects and with a sample of 808 individuals representing diverse demographics, the findings shed light on the significant influence of family structure, particularly the prevalence of extended family arrangements, on an individual's character development. Furthermore, the study underscores the pivotal role of personal characteristics, particularly self-esteem, in shaping admirable traits. The research identifies several contributing factors to positive character development, including active participation in sports, parental support, accessibility to sports facilities, positive peer influence, and high self-esteem. Parents play a crucial role in mitigating the adverse effects of peer pressure by offering positive reinforcement and serving as role models. These findings hold important implications for youth development programs, emphasizing the vital role of parents in guiding character development, particularly in the context of peer pressure.

Microscopic Mechanism for Further NO Heterogeneous Reduction by Potassium-Doped Biochar: A DFT Study.

Biomass reburning is an efficient and low-cost way to control nitric oxide (NO), and the abundant potassium (K) element in biomass affects the heterogeneous reaction between NO and biochar. Due to the incomplete simulation of the NO heterogeneous reduction reaction pathway at the molecular level and the unclear catalytic effect of K element in biochar, further research is needed on the possible next reaction and the influencing mechanism of the element. After the products of the existing reaction pathways are referenced, two reasonably simplified biochar structural models are selected as the basic reactants to study the microscopic mechanism for further NO heterogeneous reduction on the biochar surface before and after doping with the K atom based on density functional theory. In studying the two further NO heterogeneous reduction reaction pathways, we find that the carbon monoxide (CO) molecule fragment protrudes from the surface of biochar models with the desorption of N2 at the TS4 transition state, and the two edge types of biochar product models obtained by simulation calculation are Klein edge and ac56 edge observed in the experiment. In studying the catalytic effect of potassium in biochar, we find that the presence of K increases the heat release of adsorption of NO molecules, reduces the energy barrier of the rate-determining step in the nitrogen (N2) generation and desorption process (by 50.88 and 69.97%), and hinders the CO molecule from desorbing from the biochar model surface. Thermodynamic and kinetic analyses also confirm its influence. The study proves that the heterogeneous reduction reaction of four NO molecules on the surface of biochar completes the whole reaction process and provides a basic theoretical basis for the emission of nitrogen oxides (NOx) during biomass reburning.

Identification of a DNA-methylome-based signature for prognosis prediction in driver gene-negative lung adenocarcinoma.

"Driver gene-negative" lung adenocarcinoma (LUAD) was of rare treatment options and a poor prognosis. Presently, for them, few biomarkers are available for stratification analysis to make appropriate treatment strategy. This study aimed to develop a DNA-methylome-based signature to realize the precise risk-stratifying. Here, an Illumina MethylationEPIC Beadchip was applied to obtain differentially methylated CpG sites (DMCs). A four-CpG-based signature, named as TLA, was successfully established, whose prognosis-predicting power was well verified in one internal (n = 78) and other external (n = 110) validation cohorts. Patients with high-risk scores had shorter overall survival (OS) in all cohorts [hazard ratio (HR): 11.79, 5.16 and 2.99, respectively]. Additionally, it can effectively divide patients into low-risk and high-risk groups, with significantly different OS in the diverse subgroups stratified by the standard clinical parameters. As an independent prognostic factor, TLA may assist in improving the nomogram's 5-year OS-predicting ability (AUC 0.756, 95% CI:0.695-0.816), superior to TNM alone (AUC 0.644, 95% CI: 0.590-0.698). Additionally, the relationship of TLA-related genes, TAC1, LHX9, and ALX1, with prognosis and tumour invasion made them serve as potential therapy targets for driver gene-negative LUAD.

Rigid tape dynamic fixation in conservative treatment of acute anterior talofibular ligament tear: A retrospective cohort study.

Objectives: This study aims to compare results of rigid tape (RT) dynamic fixation and static fixation in conservative treatment of acute anterior talofibular ligament (ATFL) tear. Patients and methods: Between September 2021 and December 2021, a total of 91 patients (41 males, 50 females, mean age: 28.5±6.5 years, range, 18 to 40 years) who were diagnosed with ATFL tear and underwent rigid tape (RT) or cast/brace rehabilitation protocol were retrospectively analyzed. The patients were divided into two groups as the RT group (n=36) and the control group (n=55). Follow-up (FU) was performed at six months. Outcomes included pain (Numerical Rating Scale [NRS]), ankle function (American Orthopaedic Foot & Ankle Society [AOFAS] hindfoot score), deviation of center of gravity (DCG), and symptoms after returning to sports. Results: The difference at each time point of pain, AOFAS, DCG and SRS between the two groups was statistically significant (p<0.05 for all). Only one patient at Week 12 in the RT group had pain in the lateral side of the ankle, while 36 patients at Week 12 and 21 patients (18 in the medial side) at FU had pain in the control group. Conclusion: Our study results suggest that RT dynamic fixation can accurately lock the ATFL function and may prevent pseudo-stability, so as to quickly repair injury, restore function, and return to sports earlier.

Research on helmet wearing detection method based on deep learning.

The vigorous development of the construction industry has also brought unprecedented safety risks. The wearing of safety helmets at the construction site can effectively reduce casualties. As a result, this paper suggests employing a deep learning-based approach for the real-time detection of safety helmet usage among construction workers. Based on the selected YOLOv5s network through experiments, this paper analyzes its training results. Considering its poor detection effect on small objects and occluded objects. Therefore, multiple attention mechanisms are used to improve the YOLOv5s network, the feature pyramid network is improved into a BiFPN bidirectional feature pyramid network, and the post-processing method NMS is improved into Soft-NMS. Based on the above-improved method, the loss function is improved to enhance the convergence speed of the model and improve the detection speed. We propose a network model called BiFEL-YOLOv5s, which combines the BiFPN network and Focal-EIoU Loss to improve YOLOv5s. The average precision of the model is increased by 0.9% the recall rate is increased by 2.8%, and the detection speed of the model does not decrease too much. It is better suited for real-time safety helmet object detection, addressing the requirements of helmet detection across various work scenarios.

New insights into the effect of pyrolysis temperature on the spectroscopy properties of dissolved organic matter in manure-based biochar.

Dissolved organic matter (DOM) derived from biochar takes a crucial role in transport and bioavailability toward contaminants; hence, it is undeniable that a thorough analysis of its properties is important. So far, the effect of pyrolysis temperature on the functional groups, components, and evolutionary sequence of manure-based biochar DOM has not been adequately investigated. Here, DOM was released from two typical livestock manures (cow and pig) at five pyrolysis temperatures (300 ~ 700°C), and it was explored in depth with the aid of moving window 2D correlation spectroscopy (MW-2D-COS) and heterogeneous 2D correlation spectroscopy (hetero-2D-COS). The results demonstrated that the concentration, aromaticity, and hydrophobicity of DOM were greater at high temperatures, and more DOM was liberated from cow manure-based biochar at identical temperature. Protein-like compounds dominated at high temperatures. The pyrolysis temperatures of final configuration transformation points of the fulvic acid-like component and the aromatic ring C=C in DOM were 400°C and 500°C, respectively. Moreover, Fourier transform infrared spectroscopy combined with two-dimensional correlation analysis indicated that the functional group evolution of DOM depends on the pyrolysis temperature and feedstock type. The study provides a new perspective on manure management and environmental applications of biochar.

Vegetation cover change and its response to climate extremes in the Yellow River Basin.

Extreme climate events have increased in frequency and severity under the background of climate change, with vegetation growth exhibiting a sensitive response to them. By assimilating GIMMS NDVI and MODIS NDVI using the Residual Network, we obtained a long time series and high resolution NDVI dataset of the Yellow River Basin (YRB). The dataset was utilized for examining the spatiotemporal variability of NDVI and analyzing the response of vegetation cover to climate extremes with meteorological data. Our findings reveal the following: (1) A significant rise in NDVI was seen in the YRB, displaying a mean growth rate of 0.019/10a (p < 0.001). However, seasonal differences exist. The mean NDVI of multi-year declines from southeast to northwest, while the overall trend of vegetation cover improves. (2) The NDVI response to extreme temperature exhibits noticeable spatiotemporal differences. Daytime extreme high temperature in the northern YRB is negatively correlated with NDVI, while they are positively correlated in the lower YRB and the southern part of the middle YRB. Nighttime extreme high temperature exhibits a positive correlation with NDVI. Overall, NDVI displays a stronger response to extreme precipitation than to extreme temperature, with a negative correlation with CWD and a positive correlation with PRCPTOT. (3) The NDVI demonstrates a lagged response to climate extremes in the YRB, with a greater lag in response to extreme temperature than extreme precipitation. The research findings can provide scientific support for the future management and planning of vegetation in the YRB, as well as contribute to the promotion of ecological environment regulation and sustainable development.

Comprehensive analysis of CXCL14 uncovers its role during liver metastasis in colon cancer.

BACKGROUND: The most common cause of death for colon cancer patients is liver metastasis. METHODS: All the data enrolled in this study were downloaded from two public databases, The Cancer Genome Atlas Program, the TCGA-COAD project and Gene Expression Omnibus, GSE41258 project. All the analysis was performed in R software. RESULTS: In our study, we systematically explored the molecules involved in the liver metastasis process of colon cancer. The biological role of these molecules was identified through the GO and KEGG analysis. Moreover, we identified that the molecules SERPINA3, SERPINA1, MMP3, ALDH1A3, PBK and CXCL14 were the independent factors for patients survival. The CXCL14 was selected for further analysis for its most significant P value. Single-cell analysis showed that the CXCL14 was mainly expressed in the fibroblasts. Meanwhile, the biological role of fibroblasts in the colon cancer microenvironment was investigated. Further, the clinical role of CXCL14 in colon cancer was also explored. The result showed that the CXCL14 is a protective factor against colon cancer independent of other clinical parameters like age, gender, clinical stage, and TNM classifications. Then, biological enrichment analysis indicated that the CXCL14 is predominantly involved in the activating of the WNT/ß/catenin pathway, pancreas beta cells, peroxisome and bile acid metabolism. Immune infiltration analysis showed that for the patients with high CXCL14 levels, the plasma B cells, CD8 + T cells, neutrophil and NK cells might infiltrate more, in contrast to B cells, monocyte and macrophages. Furthermore, we found that the patients with low CXCL14 expression might be more sensitive to etoposide, rapamycin and sunitinib. CONCLUSIONS: Our result could improve the understanding of the liver metastasis process in colon cancer. Also, CXCL14 was identified as an underlying therapeutic target for colon cancer.

A Rare Case of Blastic Plasmacytoid Dendritic Cell Neoplasm Occurred in Postchemotherapy of Breast Cancer.

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare and highly aggressive hematologic malignancy that arises from plasmacytoid dendritic cells. BPDCN typically presents with skin lesions and may involve peripheral blood, bone marrow, lymph nodes, or extranodal sites. It usually arises de novo, and some BPDCN cases are associated with or develop into myeloid neoplasms. Here, we report a case of a 57-year-old female presenting with cervical lymphadenopathy and skin rashes during the COVID-19 pandemic in 2021 following multiple types of postmastectomy therapy for breast cancer. The patient was ultimately diagnosed with BPCDN by lymph node biopsy. To the best of our knowledge, this is the first case report of BPDCN occurring postchemotherapy of breast cancer.

Modified UiO-66-Br Microphotocatalyst with High Electron Mobility Enhances Tetracycline Degradation.

In this work, the Br functional group on the ligand UiO-66-Br was modified with a Bi-O bond through the secondary solvothermal method, and the synthesis method of visible light catalyst UB (UiO-66-BiOBr) with high electron mobility was explored. The findings indicate that the effective charge transfer of the functional group-modified material UB is 2.98 times and 1.22 times that of BiOBr and traditional UiO-66/BiOBr heterojunctions, respectively. Under simulated sunlight irradiation, the removal rate of tetracycline can reach 88.71%, and the photocatalytic performance is 22.73 times higher than that of UiO-66-Br. Moreover, it maintains good adsorption and photocatalytic performance under different laboratory and actual engineering water environment conditions. In the complex water environment of municipal wastewater, the degradation effect reaches more than 80%. Finally, the decomposition pathways of TC and ecotoxicities of the intermediates were analyzed via combining theoretical calculation, LC-MS/MS, and T.E.S.T.

Cellulose Nanocrystal Films with NIR-II Circularly Polarized Light for Cancer Detection Applications.

Near-infrared circularly polarized light is attractive for wide-ranging applications. However, high-performance near-infrared circularly polarized light is challenging to realize. Here, we show that left-handed chiral photonic cellulose nanocrystal (CNC) films produced from ultrasonicated suspensions enable right-handed circularly polarized luminescence with a dissymmetry factor of -0.330 in the second near-infrared window (NIR-II). We present a theoretical analysis of the adverse effect of structural defects and luminescence intensity heterogeneity on the right-handed circularly polarized luminescence glum inside the bandgap and the occurrence of left-handed circularly polarized luminescence at the band edges. We demonstrate the potential of the chiral photonic CNC films with NIR-II circularly polarized light for cancer cell discrimination. The present work identifies key scientific questions in CNC-based circularly polarized luminescence materials research.

Exosomal hsa_circRNA_047733 integrated with clinical features for preoperative prediction of lymph node metastasis risk in oral squamous cell carcinoma.

BACKGROUND: To evaluate whether exosomal circRNAs could serve as diagnostic biomarkers for the accurate preoperative prediction of lymph node metastasis (LNM) risk in oral squamous cell carcinoma (OSCC) patients. METHODS: A combinative strategy of exosomal circRNAs microarray and qRT-PCR verification was employed to dig LNM-related circRNA signatures. Then, a dynamic nomogram was developed based on candidate circRNAs and preoperative clinical features and the calibration, discrimination, and clinical use of the nomogram were evaluated. RESULTS: According to the microarray, three circRNAs derived from the tumor were associated with preoperative LNM risk, including hsa_circRNA_047733, hsa_circRNA_024144 and hsa_circRNA_403472. The hsa_circRNA_047733 was further verified to be significantly downregulated in patients with LNM (+) as compared with those with LNM (-) (p = 0.007). Patients with the higher expression of hsa_circRNA_047733 showed a lower risk of LNM (multivariate-adjusted OR = 0.22, 95%CI: 0.06-0.83). The bioinformatics prediction showed that hsa_circRNA_047733 might sponge miR-4464/miR-4748 to regulate RPS21 expression. A dynamic nomogram integrating exosomal hsa_circRNA_047733 with five clinicopathological characteristics (tumor site, leukocyte level, maximum tumor diameter, and LNM reported by MRI and preoperative biopsy differentiation) was developed. The model displayed an excellent discrimination ability (AUC = 0.868, 95%CI: 0.781-0.955) and great calibration. The decision curve revealed a higher net benefit superior to the baseline model at an 80% threshold probability. CONCLUSION: The data provide preliminary evidence that exosomal hsa_circRNA_047733 might be a novel biomarker for the LNM of OSCC. The hsa_circRNA_047733-based dynamic nomogram could serve as a convenient preoperative assessment tool to predict the risk of LNM for OSCC patients.

Hexachloropentadiene in Soil, Air, and Biota Around an Agrochemical Factory: Concentrations, Distribution, and Risk Evaluation.

Hexachloropentadiene (HCPD) is a highly toxic compound that is mainly used for preparation of organochlorine insecticides. To investigate HCPD contamination of the environment during pesticide processing, 153 air, soil, and biota samples were collected around an agrochemical factory in different seasons of 1 year and analyzed for HCPD. The HCPD concentrations were 0.01-12.7 ng/m3 (average 2.60 ng/m3) in the air samples and 0.14-51.5 ng/g (average 4.11 ng/g) in the soil samples. HCPD concentrations were highest within 1 km north of the production site, which was in the downwind direction of the factory and storage tanks, especially in autumn and winter. Soil-air exchange analysis showed that HCPD was deposited from air to soil with a flux of 0.003 to 0.20 ng/(m2 d) throughout the year. The dismantling of obsolete equipment accelerated the release of HCPD into the air and increased the amount of HCPD deposited in the soil. HPCD concentration ranges were 0.44-55.7 ng/g dry weight [d.w.] (average 22.2 ng/g d.w.) and 6.69-91.4 ng/g d.w. (average 26.2) in locally grown rice and wheat, respectively. The concentration range was 12.1-1596 ng/g lipid weight (average 560 ng/g lipid weight) in local organisms, except for chicken. In tissues from locally raised chicken, the HCPD concentrations decreased in the order of gizzard, liver, heart, and meat. HCPD was amplified through a short food chain (soil, Vigna unguiculata leaves, larvae of Pieris rapae, and chicken), and the bioaccumulation factor gradually increased over a range of 1.19-25.1 (mean 9.81).

U-Shaped Relationship of Rare Earth Element Lanthanum and Oral Cancer Risk: A Propensity Score-Based Study in the Southeast of China.

As an important rare earth element (REE) extensively applied to industry, agriculture, and medicine, lanthanum (La) has attracted a host of health concerns. This study aimed to explore the relationship between La exposure and the risk of developing oral cancer through a case-control study with a large sample size. Serum La levels of 430 oral cancer patients and 1,118 healthy controls were detected by inductively coupled plasma mass spectrometry (ICP-MS). The association of La level with the risk of oral cancer was assessed in two ways: (1) as a continuous scale based on restricted cubic splines (RCS); (2) as a priori defined centile categories using multivariate logistic regression model, based on propensity score matching (PSM) and inverse probability of treatment weighting (IPTW). The RCS revealed a non-linear U-shaped relationship between serum La and oral cancer risk. Serum La deficiency or excess was associated with an increased risk of oral cancer. When the La level was analyzed as a categorical variable, a similar U-shaped association was observed. Of note, compared to those with La concentrations of 0.243-0.341 µg/L (reference quantiles, 41st-60th), the risk was increased in those with the lower or higher quantiles (0.132-0.242 µg/L vs. 0.243-0.341 µg/L: OR = 1.80, 95%CI: 1.07-3.02; 0.342-0.497 µg/L vs. 0.243-0.341 µg/L: OR = 2.30, 95%CI: 1.38-3.84). The results were generally consistent with the PSM and IPTW analyses. This preliminary study provides strong evidence that there was a U-shaped relationship between serum La levels and oral cancer risk. Much additional work is warranted to confirm our findings.

Carboxylated cellulose nanocrystal films with tunable chiroptical properties.

Carboxylated cellulose nanocrystals prepared by TEMPO-mediated oxidation exhibit a distinct ability to form nematic order, however, their ability to form chiral nematic films remains relatively unexplored. In this study, bleached cotton pulp hydrolyzed with hydrochloric acid and oxidized by TEMPO-mediated oxidation produce carboxylated cellulose nanocrystals with different aspect ratios 33.1, 32.8, 30.9, 29.0 and 28.9, and surface charge densities 0.16, 0.56, 1.00, 1.25, and 1.42 e·nm-2. By tuning the aspect ratio and surface charge density, the optimal carboxylated cellulose nanocrystals producing left-handed chiral nematic films by evaporation-induced self-assembly are obtained. The left-handed chiral nematic films enable selective reflection of left-handed circularly polarized light with the peak wavelength tunable from the visible to the near-infrared regime by modifying the characteristics of nanorods and suspensions. Such carboxylated cellulose nanocrystal films transform spontaneous luminescence to right-handed circularly polarized luminescence with the peak luminescence dissymmetry factor of -0.51.

Deep learning for evaluation of microvascular invasion in hepatocellular carcinoma from tumor areas of histology images.

BACKGROUND: Microvascular invasion (MVI) is essential for the management of hepatocellular carcinoma (HCC). However, MVI is hard to evaluate in patients without sufficient peri-tumoral tissue samples, which account for over a half of HCC patients. METHODS: We established an MVI deep-learning (MVI-DL) model with a weakly supervised multiple-instance learning framework, to evaluate MVI status using only tumor tissues from the histological whole slide images (WSIs). A total of 350 HCC patients (2917 WSIs) from the First Affiliated Hospital of Sun Yat-sen University (FAHSYSU cohort) were divided into a training and test set. One hundred and twenty patients (504 WSIs) from Dongguan People's Hospital and Shunde Hospital of Southern Medical University (DG-SD cohort) formed an external test set. Unsupervised clustering and class activation mapping were applied to visualize the key histological features. RESULTS: In the FAHSYSU and DG-SD test set, the MVI-DL model achieved an AUC of 0.904 (95% CI 0.888-0.920) and 0.871 (95% CI 0.837-0.905), respectively. Visualization results showed that macrotrabecular architecture with rich blood sinus, rich tumor stroma and high intratumor heterogeneity were identified as the key features associated with MVI ( +), whereas severe immune infiltration and highly differentiated tumor cells were associated with MVI (-). In the simulation of patients with only one WSI or biopsies only, the AUC of the MVI-DL model reached 0.875 (95% CI 0.855-0.895) and 0.879 (95% CI 0.853-0.906), respectively. CONCLUSION: The effective, interpretable MVI-DL model has potential as an important tool with practical clinical applicability in evaluating MVI status from the tumor areas on the histological slides.

Innovative Synoptic Reporting With Seven-Point Sampling Protocol to Improve Detection Rate of Microvascular Invasion in Hepatocellular Carcinoma.

Pathological MVI diagnosis could help to determine the prognosis and need for adjuvant therapy in hepatocellular carcinoma (HCC). However, narrative reporting (NR) would miss relevant clinical information and non-standardized sampling would underestimate MVI detection. Our objective was to explore the impact of innovative synoptic reporting (SR) and seven-point sampling (SPRING) protocol on microvascular invasion (MVI) rate and patient outcomes. In retrospective cohort, we extracted MVI status from NR in three centers and re-reviewed specimen sections by SR recommended by the College of American Pathologists (CAP) in our center. In prospective cohort, our center implemented the SPRING protocol, and external centers remained traditional pathological examination. MVI rate was compared between our center and external centers in both cohorts. Recurrence-free survival (RFS) before and after implementation was calculated by Kaplan-Meier method and compared by the log-rank test. In retrospective study, we found there was no significant difference in MVI rate between our center and external centers [10.3% (115/1112) vs. 12.4% (35/282), P=0.316]. In our center, SR recommended by CAP improved the MVI detection rate from 10.3 to 38.6% (P<0.001). In prospective study, the MVI rate in our center under SPRING was significantly higher than external centers (53.2 vs. 17%, P<0.001). RFS of MVI (-) patients improved after SPRING in our center (P=0.010), but it remained unchanged in MVI (+) patients (P=0.200). We conclude that the SR recommended by CAP could help to improve MVI detection rate. Our SPRING protocol could help to further improve the MVI rate and optimize prognostic stratification for HCC patients.

Multi-Parametric MRI-Based Radiomics Models for Predicting Molecular Subtype and Androgen Receptor Expression in Breast Cancer.

OBJECTIVE: To investigate whether radiomics features extracted from multi-parametric MRI combining machine learning approach can predict molecular subtype and androgen receptor (AR) expression of breast cancer in a non-invasive way. MATERIALS AND METHODS: Patients diagnosed with clinical T2-4 stage breast cancer from March 2016 to July 2020 were retrospectively enrolled. The molecular subtypes and AR expression in pre-treatment biopsy specimens were assessed. A total of 4,198 radiomics features were extracted from the pre-biopsy multi-parametric MRI (including dynamic contrast-enhancement T1-weighted images, fat-suppressed T2-weighted images, and apparent diffusion coefficient map) of each patient. We applied several feature selection strategies including the least absolute shrinkage and selection operator (LASSO), and recursive feature elimination (RFE), the maximum relevance minimum redundancy (mRMR), Boruta and Pearson correlation analysis, to select the most optimal features. We then built 120 diagnostic models using distinct classification algorithms and feature sets divided by MRI sequences and selection strategies to predict molecular subtype and AR expression of breast cancer in the testing dataset of leave-one-out cross-validation (LOOCV). The performances of binary classification models were assessed via the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). And the performances of multiclass classification models were assessed via AUC, overall accuracy, precision, recall rate, and F1-score. RESULTS: A total of 162 patients (mean age, 46.91 ± 10.08 years) were enrolled in this study; 30 were low-AR expression and 132 were high-AR expression. HR+/HER2- cancers were diagnosed in 56 cases (34.6%), HER2+ cancers in 81 cases (50.0%), and TNBC in 25 patients (15.4%). There was no significant difference in clinicopathologic characteristics between low-AR and high-AR groups (P > 0.05), except the menopausal status, ER, PR, HER2, and Ki-67 index (P = 0.043, <0.001, <0.001, 0.015, and 0.006, respectively). No significant difference in clinicopathologic characteristics was observed among three molecular subtypes except the AR status and Ki-67 (P = <0.001 and 0.012, respectively). The Multilayer Perceptron (MLP) showed the best performance in discriminating AR expression, with an AUC of 0.907 and an accuracy of 85.8% in the testing dataset. The highest performances were obtained for discriminating TNBC vs. non-TNBC (AUC: 0.965, accuracy: 92.6%), HER2+ vs. HER2- (AUC: 0.840, accuracy: 79.0%), and HR+/HER2- vs. others (AUC: 0.860, accuracy: 82.1%) using MLP as well. The micro-AUC of MLP multiclass classification model was 0.896, and the overall accuracy was 0.735. CONCLUSIONS: Multi-parametric MRI-based radiomics combining with machine learning approaches provide a promising method to predict the molecular subtype and AR expression of breast cancer non-invasively.