Applying dynamic contrast-enhanced MRI tracer kinetic models to differentiate benign and malignant soft tissue tumors.

BACKGROUND: To explore the potential of different quantitative dynamic contrast-enhanced (qDCE)-MRI tracer kinetic (TK) models and qDCE parameters in discriminating benign from malignant soft tissue tumors (STTs). METHODS: This research included 92 patients (41females, 51 males; age range 16-86 years, mean age 51.24 years) with STTs. The qDCE parameters (Ktrans, Kep, Ve, Vp, F, PS, MTT and E) for regions of interest of STTs were estimated by using the following TK models: Tofts (TOFTS), Extended Tofts (EXTOFTS), adiabatic tissue homogeneity (ATH), conventional compartmental (CC), and distributed parameter (DP). We established a comprehensive model combining the morphologic features, time-signal intensity curve shape, and optimal qDCE parameters. The capacities to identify benign and malignant STTs was evaluated using the area under the curve (AUC), degree of accuracy, and the analysis of the decision curve. RESULTS: TOFTS-Ktrans, EXTOFTS-Ktrans, EXTOFTS-Vp, CC-Vp and DP-Vp demonstrated good diagnostic performance among the qDCE parameters. Compared with the other TK models, the DP model has a higher AUC and a greater level of accuracy. The comprehensive model (AUC, 0.936, 0.884-0.988) demonstrated superiority in discriminating benign and malignant STTs, outperforming the qDCE models (AUC, 0.899-0.915) and the traditional imaging model (AUC, 0.802, 0.712-0.891) alone. CONCLUSIONS: Various TK models successfully distinguish benign from malignant STTs. The comprehensive model is a noninvasive approach incorporating morphological imaging aspects and qDCE parameters, and shows significant potential for further development.

Comprehensive characterization of IFNγ signaling in acute myeloid leukemia reveals prognostic and therapeutic strategies.

Interferon gamma (IFNγ) is a critical cytokine known for its diverse roles in immune regulation, inflammation, and tumor surveillance. However, while IFNγ levels were elevated in sera of most newly diagnosed acute myeloid leukemia (AML) patients, its complex interplay in AML remains insufficiently understood. We aim to characterize these complex interactions through comprehensive bulk and single-cell approaches in bone marrow of newly diagnosed AML patients. We identify monocytic AML as having a unique microenvironment characterized by IFNγ producing T and NK cells, high IFNγ signaling, and immunosuppressive features. IFNγ signaling score strongly correlates with venetoclax resistance in primary AML patient cells. Additionally, IFNγ treatment of primary AML patient cells increased venetoclax resistance. Lastly, a parsimonious 47-gene IFNγ score demonstrates robust prognostic value. In summary, our findings suggest that inhibiting IFNγ is a potential treatment strategy to overcoming venetoclax resistance and immune evasion in AML patients.

Paired single-B-cell transcriptomics and receptor sequencing reveal activation states and clonal signatures that characterize B cells in acute myeloid leukemia.

BACKGROUND: Acute myeloid leukemia (AML) is associated with a dismal prognosis. Immune checkpoint blockade (ICB) to induce antitumor activity in AML patients has yielded mixed results. Despite the pivotal role of B cells in antitumor immunity, a comprehensive assessment of B lymphocytes within AML's immunological microenvironment along with their interaction with ICB remains rather constrained. METHODS: We performed an extensive analysis that involved paired single-cell RNA and B-cell receptor (BCR) sequencing on 52 bone marrow aspirate samples. These samples included 6 from healthy bone marrow donors (normal), 24 from newly diagnosed AML patients (NewlyDx), and 22 from 8 relapsed or refractory AML patients (RelRef), who underwent assessment both before and after azacitidine/nivolumab treatment. RESULTS: We delineated nine distinct subtypes of B cell lineage in the bone marrow. AML patients exhibited reduced nascent B cell subgroups but increased differentiated B cells compared with healthy controls. The limited diversity of BCR profiles and extensive somatic hypermutation indicated antigen-driven affinity maturation within the tumor microenvironment of RelRef patients. We established a strong connection between the activation or stress status of naïve and memory B cells, as indicated by AP-1 activity, and their differentiation state. Remarkably, atypical memory B cells functioned as specialized antigen-presenting cells closely interacting with AML malignant cells, correlating with AML stemness and worse clinical outcomes. In the AML microenvironment, plasma cells demonstrated advanced differentiation and heightened activity. Notably, the clinical response to ICB was associated with B cell clonal expansion and plasma cell function. CONCLUSIONS: Our findings establish a comprehensive framework for profiling the phenotypic diversity of the B cell lineage in AML patients, while also assessing the implications of immunotherapy. This will serve as a valuable guide for future inquiries into AML treatment strategies.

An MRI-Based Deep Transfer Learning Radiomics Nomogram to Predict Ki-67 Proliferation Index of Meningioma.

The objective of this study was to predict Ki-67 proliferation index of meningioma by using a nomogram based on clinical, radiomics, and deep transfer learning (DTL) features. A total of 318 cases were enrolled in the study. The clinical, radiomics, and DTL features were selected to construct models. The calculation of radiomics and DTL score was completed by using selected features and correlation coefficient. The deep transfer learning radiomics (DTLR) nomogram was constructed by selected clinical features, radiomics score, and DTL score. The area under the receiver operator characteristic curve (AUC) was calculated. The models were compared by Delong test of AUCs and decision curve analysis (DCA). The features of sex, size, and peritumoral edema were selected to construct clinical model. Seven radiomics features and 15 DTL features were selected. The AUCs of clinical, radiomics, DTL model, and DTLR nomogram were 0.746, 0.75, 0.717, and 0.779 respectively. DTLR nomogram had the highest AUC of 0.779 (95% CI 0.6643-0.8943) with an accuracy rate of 0.734, a sensitivity value of 0.719, and a specificity value of 0.75 in test set. There was no significant difference in AUCs among four models in Delong test. The DTLR nomogram had a larger net benefit than other models across all the threshold probability. The DTLR nomogram had a satisfactory performance in Ki-67 prediction and could be a new evaluation method of meningioma which would be useful in the clinical decision-making.

Deep Learning Radiomics Nomogram Based on Enhanced CT to Predict the Response of Metastatic Lymph Nodes to Neoadjuvant Chemotherapy in Locally Advanced Gastric Cancer.

BACKGROUND: We aimed to construct and validate a deep learning (DL) radiomics nomogram using baseline and restage enhanced computed tomography (CT) images and clinical characteristics to predict the response of metastatic lymph nodes to neoadjuvant chemotherapy (NACT) in locally advanced gastric cancer (LAGC). METHODS: We prospectively enrolled 112 patients with LAGC who received NACT from January 2021 to August 2022. After applying the inclusion and exclusion criteria, 98 patients were randomized 7:3 to the training cohort (n = 68) and validation cohort (n = 30). We established and compared three radiomics signatures based on three phases of CT images before and after NACT, namely radiomics-baseline, radiomics-delta, and radiomics-restage. Then, we developed a clinical model, DL model, and a nomogram to predict the response of LAGC after NACT. We evaluated the predictive accuracy and clinical validity of each model using the receiver operating characteristic curve and decision curve analysis, respectively. RESULTS: The radiomics-delta signature was the best predictor among the three radiomics signatures. So, we developed and validated a DL delta radiomics nomogram (DLDRN). In the validation cohort, the DLDRN produced an area under the receiver operating curve of 0.94 (95% confidence interval, 0.82-0.96) and demonstrated adequate differentiation of good response to NACT. Furthermore, the DLDRN significantly outperformed the clinical model and DL model (p < 0.001). The clinical utility of the DLDRN was confirmed through decision curve analysis. CONCLUSIONS: In patients with LAGC, the DLDRN effectively predicted a therapeutic response in metastatic lymph nodes, which could provide valuable information for individualized treatment.

Na Cocations and Hydrothermal Aging Cooperatively Boost the Regeneration of Phosphorus-Poisoned Pd/SSZ-13 for Passive NOx Adsorption.

Pd/SSZ-13 has been proposed as a passive NOx adsorber (PNA) for low-temperature NOx adsorption. However, it remains challenging for Pd/SSZ-13 to work efficiently when suffering from phosphorus poisoning. Herein, we report a simple and efficient strategy to regenerate the phosphorus-poisoned Pd/SSZ-13 based on the cooperation between hydrothermal aging treatment and Na cocations. It was found that hydrothermal aging treatment enabled the redispersion of Pd and P-containing species in phosphorus-poisoned Pd/SSZ-13. Meanwhile, the presence of Na cocations significantly reduced the formation of AlPO4 and retained more paired Al sites for highly dispersed Pd2+ ions, which was of great importance for the recovery of adsorption performance. To our satisfaction, the restoration ratio of the adsorption capacity of poisoned Pd/SSZ-13 was >90% after regeneration. Strikingly, the NOx adsorption activities of phosphorus-poisoned Pd/SSZ-13 with phosphorus loadings of 0.2 and 0.4 mmol g-1 almost completely recovered upon regeneration. This study demonstrates the promoting effect of Na cocations on the regeneration of phosphorus-poisoned Pd/SSZ-13 by hydrothermal aging treatment, which provides useful guidance for the design of PNA materials with excellent durability for cold-start application.

Single-cell chromatin accessibility profiling of acute myeloid leukemia reveals heterogeneous lineage composition upon therapy-resistance.

Acute myeloid leukemia (AML) is a heterogeneous disease characterized by high rate of therapy resistance. Since the cell of origin can impact response to therapy, it is crucial to understand the lineage composition of AML cells at time of therapy resistance. Here we leverage single-cell chromatin accessibility profiling of 22 AML bone marrow aspirates from eight patients at time of therapy resistance and following subsequent therapy to characterize their lineage landscape. Our findings reveal a complex lineage architecture of therapy-resistant AML cells that are primed for stem and progenitor lineages and spanning quiescent, activated and late stem cell/progenitor states. Remarkably, therapy-resistant AML cells are also composed of cells primed for differentiated myeloid, erythroid and even lymphoid lineages. The heterogeneous lineage composition persists following subsequent therapy, with early progenitor-driven features marking unfavorable prognosis in The Cancer Genome Atlas AML cohort. Pseudotime analysis further confirms the vast degree of heterogeneity driven by the dynamic changes in chromatin accessibility. Our findings suggest that therapy-resistant AML cells are characterized not only by stem and progenitor states, but also by a continuum of differentiated cellular lineages. The heterogeneity in lineages likely contributes to their therapy resistance by harboring different degrees of lineage-specific susceptibilities to therapy.

Evolution of immune and stromal cell states and ecotypes during gastric adenocarcinoma progression.

Understanding tumor microenvironment (TME) reprogramming in gastric adenocarcinoma (GAC) progression may uncover novel therapeutic targets. Here, we performed single-cell profiling of precancerous lesions, localized and metastatic GACs, identifying alterations in TME cell states and compositions as GAC progresses. Abundant IgA+ plasma cells exist in the premalignant microenvironment, whereas immunosuppressive myeloid and stromal subsets dominate late-stage GACs. We identified six TME ecotypes (EC1-6). EC1 is exclusive to blood, while EC4, EC5, and EC2 are highly enriched in uninvolved tissues, premalignant lesions, and metastases, respectively. EC3 and EC6, two distinct ecotypes in primary GACs, associate with histopathological and genomic characteristics, and survival outcomes. Extensive stromal remodeling occurs in GAC progression. High SDC2 expression in cancer-associated fibroblasts (CAFs) is linked to aggressive phenotypes and poor survival, and SDC2 overexpression in CAFs contributes to tumor growth. Our study provides a high-resolution GAC TME atlas and underscores potential targets for further investigation.

Proteogenomic landscape of gastric adenocarcinoma peritoneal metastases.

Advanced gastric adenocarcinoma (GAC) often leads to peritoneal carcinomatosis (PC) and is associated with very poor outcome. Here we report the comprehensive proteogenomic study of ascites derived cells from a prospective GAC cohort (n = 26 patients with peritoneal carcinomatosis, PC). A total of 16,449 proteins were detected from whole cell extracts (TCEs). Unsupervised hierarchical clustering resulted in three distinct groups that reflected extent of enrichment in tumor cells. Integrated analysis revealed enriched biological pathways and notably, some druggable targets (cancer-testis antigens, kinases, and receptors) that could be exploited to develop effective therapies and/or tumor stratifications. Systematic comparison of expression levels of proteins and mRNAs revealed special expression patterns of key therapeutics target notably high mRNA and low protein expression of HAVCR2 (TIM-3), and low mRNA but high protein expression of cancer-testis antigens CTAGE1 and CTNNA2. These results inform strategies to target GAC vulnerabilities.

Single-Cell Profiling of CD8+ T Cells in Acute Myeloid Leukemia Reveals a Continuous Spectrum of Differentiation and Clonal Hyperexpansion.

Comprehensive investigation of CD8+ T cells in acute myeloid leukemia (AML) is essential for developing immunotherapeutic strategies beyond immune checkpoint blockade. Herein, we performed single-cell RNA profiling of CD8+ T cells from 3 healthy bone marrow donors and 23 newly diagnosed (NewlyDx) and 8 relapsed/refractory (RelRef) patients with AML. Cells coexpressing canonical exhaustion markers formed a cluster constituting <1% of all CD8+ T cells. We identified two effector CD8+ T-cell subsets characterized by distinct cytokine and metabolic profiles that were differentially enriched in NewlyDx and RelRef patients. We refined a 25-gene CD8-derived signature correlating with therapy resistance, including genes associated with activation, chemoresistance, and terminal differentiation. Pseudotemporal trajectory analysis supported enrichment of a terminally differentiated state in CD8+ T cells with high CD8-derived signature expression at relapse or refractory disease. Higher expression of the 25-gene CD8 AML signature correlated with poorer outcomes in previously untreated patients with AML, suggesting that the bona fide state of CD8+ T cells and their degree of differentiation are clinically relevant. Immune clonotype tracking revealed more phenotypic transitions in CD8 clonotypes in NewlyDx than in RelRef patients. Furthermore, CD8+ T cells from RelRef patients had a higher degree of clonal hyperexpansion associated with terminal differentiation and higher CD8-derived signature expression. Clonotype-derived antigen prediction revealed that most previously unreported clonotypes were patient-specific, suggesting significant heterogeneity in AML immunogenicity. Thus, immunologic reconstitution in AML is likely to be most successful at earlier disease stages when CD8+ T cells are less differentiated and have greater capacity for clonotype transitions.

Single cell clonotypic and transcriptional evolution of multiple myeloma precursor disease.

Multiple myeloma remains an incurable disease, and the cellular and molecular evolution from precursor conditions, including monoclonal gammopathy of undetermined significance and smoldering multiple myeloma, is incompletely understood. Here, we combine single-cell RNA and B cell receptor sequencing from fifty-two patients with myeloma precursors in comparison with myeloma and normal donors. Our comprehensive analysis reveals early genomic drivers of malignant transformation, distinct transcriptional features, and divergent clonal expansion in hyperdiploid versus non-hyperdiploid samples. Additionally, we observe intra-patient heterogeneity with potential therapeutic implications and identify distinct patterns of evolution from myeloma precursor disease to myeloma. We also demonstrate distinctive characteristics of the microenvironment associated with specific genomic changes in myeloma cells. These findings add to our knowledge about myeloma precursor disease progression, providing valuable insights into patient risk stratification, biomarker discovery, and possible clinical applications.

Metabolic modulation of CtBP dimeric status impacts the repression of DNA damage repair genes and the platinum sensitivity of ovarian cancer.

Platinum drug-based chemotherapy plays a dominant role in OC (ovarian cancer) treatment. The expression of DNA damage repair (DDR) genes is critical in distinguishing drug-sensitive and drug-refractory patients, as well as in the development of drug resistance in long-term treated patients. CtBP is a highly expressed oncogene in OC and was found to repress DDR genes expression in our previous study. In the present study, the formation of CtBP dimers in live cells was studied, and the functional differences between monomeric and oligomeric CtBP were explored by CHIP-seq and RNA-seq. Besides, the dynamics of CtBP dimer formation in response to the metabolic modulation were investigated by the protein fragment complementation (PCA) assays. We show that dimerized CtBP, but not the dimerization-defective mutant, binds to and represses DDR gene expression in OC cells. Treatment of the mice tumors grown from engrafted OC cells by cisplatin disclosed that high-level CtBP expression promotes the CtBP dimerization and increases the therapeutic effect of cisplatin. Moreover, the CtBP dimerization is responsive to the intracellular metabolic status as represented by the free NADH abundance. Metformin was found to increase the dimerization of CtBP and potentiate the therapeutic effect of cisplatin in a CtBP dimerization-dependent manner. Our data suggest that the CtBP dimerization status is a potential biomarker to predict platinum drug sensitivity in patients with ovarian cancer and a target of metformin to improve the therapeutic effect of platinum drugs in OC treatment.

Single-Cell Profiling of CD8+ T Cells in Acute Myeloid Leukemia Reveals a Continuous Spectrum of Differentiation and Clonal Hyperexpansion.

Comprehensive investigation of CD8+ T cells in acute myeloid leukemia (AML) is essential for developing immunotherapeutic strategies beyond immune checkpoint blockade. Herein, we performed single-cell RNA profiling of CD8+ T cells from 3 healthy bone marrow donors and 23 newly diagnosed (NewlyDx) and 8 relapsed/refractory (RelRef) AML patients. Cells co-expressing canonical exhaustion markers formed a cluster constituting <1% of all CD8+ T cells. We identified two effector CD8+ T cell subsets characterized by distinct cytokine and metabolic profiles that were differentially enriched in NewlyDx and RelRef patients. We refined a 25-gene CD8-derived signature correlating with therapy resistance, including genes associated with activation, chemoresistance, and terminal differentiation. Pseudotemporal trajectory analysis supported enrichment of a terminally differentiated state in CD8+ T cells with high CD8-derived signature expression at relapse or refractory disease. Higher expression of the 25-gene CD8 AML signature correlated with poorer outcomes in previously untreated AML patients, suggesting that the bona fide state of CD8+ T cells and their degree of differentiation are clinically relevant. Immune clonotype tracking revealed more phenotypic transitions in CD8 clonotypes in NewlyDx than in RelRef patients. Furthermore, CD8+ T cells from RelRef patients had a higher degree of clonal hyperexpansion associated with terminal differentiation and higher CD8-derived signature expression. Clonotype-derived antigen prediction revealed that most previously unreported clonotypes were patient-specific, suggesting significant heterogeneity in AML immunogenicity. Thus, immunologic reconstitution in AML is likely to be most successful at earlier disease stages when CD8+ T cells are less differentiated and have greater capacity for clonotype transitions.

A Computed Tomography Radiomics Nomogram in Differentiating Chordoma From Giant Cell Tumor in the Axial Skeleton.

OBJECTIVE: The aim of the study is to develop and validate a computed tomography (CT) radiomics nomogram for preoperatively differentiating chordoma from giant cell tumor (GCT) in the axial skeleton. METHODS: Seventy-three chordomas and 38 GCTs in axial skeleton were retrospectively included and were divided into a training cohort (n = 63) and a test cohort (n = 48). The radiomics features were extracted from CT images. A radiomics signature was developed by using the least absolute shrinkage and selection operator model, and a radiomics score (Rad-score) was acquired. By combining the Rad-score with independent clinical risk factors using multivariate logistic regression model, a radiomics nomogram was established. Calibration and receiver operator characteristic curves were used to assess the performance of the nomogram. RESULTS: Five features were selected to construct the radiomics signature. The radiomics signature showed favorable discrimination in the training cohort (area under the curve [AUC], 0.860; 95% confidence interval [CI], 0.760-0.960) and the test cohort (AUC, 0.830; 95% CI, 0.710-0.950). Age and location were the independent clinical factors. The radiomics nomogram combining the Rad-score with independent clinical factors showed good discrimination capability in the training cohort (AUC, 0.930; 95% CI, 0.880-0.990) and the test cohort (AUC, 0.980; 95% CI, 0.940-1.000) and outperformed the radiomics signature ( z = 2.768, P = 0.006) in the test cohort. CONCLUSIONS: The CT radiomics nomogram shows good predictive efficacy in differentiating chordoma from GCT in the axial skeleton, which might facilitate clinical decision making.

Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance.

Tumor-infiltrating T cells offer a promising avenue for cancer treatment, yet their states remain to be fully characterized. Here we present a single-cell atlas of T cells from 308,048 transcriptomes across 16 cancer types, uncovering previously undescribed T cell states and heterogeneous subpopulations of follicular helper, regulatory and proliferative T cells. We identified a unique stress response state, TSTR, characterized by heat shock gene expression. TSTR cells are detectable in situ in the tumor microenvironment across various cancer types, mostly within lymphocyte aggregates or potential tertiary lymphoid structures in tumor beds or surrounding tumor edges. T cell states/compositions correlated with genomic, pathological and clinical features in 375 patients from 23 cohorts, including 171 patients who received immune checkpoint blockade therapy. We also found significantly upregulated heat shock gene expression in intratumoral CD4/CD8+ cells following immune checkpoint blockade treatment, particularly in nonresponsive tumors, suggesting a potential role of TSTR cells in immunotherapy resistance. Our well-annotated T cell reference maps, web portal and automatic alignment/annotation tool could provide valuable resources for T cell therapy optimization and biomarker discovery.

Clinico-pathological features of diabetic and non-diabetic renal diseases in type 2 diabetic patients: a retrospective study from a 10-year experience in a single center.

AIM: To compare clinical and pathological characteristics as well as prognosis between diabetic nephropathy (DN) and non-diabetic renal disease (NDRD) so as to explore potential diagnostic criteria of DN and provide some guidance for the treatment of type 2 diabetes mellitus (T2DM) patients with kidney involvement. METHODS: T2DM patients with renal impairment who underwent kidney biopsy were included in this study, who were classified into 3 groups (DN, NDRD, DN with NDRD) based on their renal pathological diagnosis. Baseline clinical characteristics as well as follow-up data were collected and analyzed among 3 groups. Logistic regression was performed to determine the best predictors for DN diagnosis. Additional 34 MN patients without diabetes were enrolled by propensity score matching method to compare serum PLA2R antibody titer and kidney outcomes between diabetic MN patients and MN alone. RESULTS: Among 365 patients with type 2 diabetes who underwent kidney biopsy, 179 (49.0%) patients were diagnosed with NDRD alone and 37 (10.1%) patients with NDRD combined DN. Risk factors for DN development in T2DM patients were longer time since diabetes diagnosis, higher level of serum creatinine, absence of hematuria and presence of diabetic retinopathy by multivariate analysis. Lower rate of proteinuria remission and higher risk of renal progression were observed in DN group compared with NDRD group. Membranous nephropathy was the most common NDRD in diabetic patients. There was no difference in serum PLA2R antibody positiveness or titer between MN patients with or without T2DM. There was lower remission rate but similar renal progression in diabetic MN when age, gender, baseline eGFR, albuminuria and IFTA score were adjusted. CONCLUSIONS: Non-diabetic renal disease is not uncommon in T2DM patients with renal impairment, which has better prognosis with proper treatment. Coexisting diabetic status does not exert negative impact on renal progression in MN patients, and immunosuppressive agents should be administered when necessary.

Harnessing Reconstructed Macrophage Modulation of Infiltration-Excluded Immune Microenvironments To Delineate Glioma Infiltrative Region.

High invasiveness of glioma produces residual glioma cells in the brain parenchyma after surgery and ultimately causes recurrence. Precise delineation of glioma infiltrative region is critical for an accurate complete resection, which is challenging. The glioma-infiltrating area constitutes infiltration-excluded immune microenvironments (I-E TIMEs), which recruits endogenous or adoptive macrophages to the invasive edge of glioma. Thus, combined with immune cell tracing technology, we provided a novel strategy for the preoperative precise definition of the glioma infiltration boundary, even satellite-like infiltration stoves. Herein, the biomimetic probe was constructed by internalizing fluorophore labeled PEGylated KMnF3 nanoparticles into bone-marrow-derived macrophages using magnetic resonance imaging (MRI)/fluorescence imaging (FI). The biomimetic probe was able to cross the blood-brain barrier and home to the orthotopic glioma infiltrates including satellite stove under MRI and FI tracing, which was validated using hematoxylin and eosin staining, indicating its excellent performance in distinguishing the margins between the glioma cell and normal tissues. This study guides the precise definition of glioma infiltration boundaries at the cellular level, including the observation of any residual glioma cells after surgery. Thus, it has the potential to guide surgery to maximize resection and predict recurrence in the clinic.

Primary bone tumor detection and classification in full-field bone radiographs via YOLO deep learning model.

OBJECTIVES: Automatic bone lesions detection and classifications present a critical challenge and are essential to support radiologists in making an accurate diagnosis of bone lesions. In this paper, we aimed to develop a novel deep learning model called You Only Look Once (YOLO) to handle detecting and classifying bone lesions on full-field radiographs with limited manual intervention. METHODS: In this retrospective study, we used 1085 bone tumor radiographs and 345 normal bone radiographs from two centers between January 2009 and December 2020 to train and test our YOLO deep learning (DL) model. The trained model detected bone lesions and then classified these radiographs into normal, benign, intermediate, or malignant types. The intersection over union (IoU) was used to assess the model's performance in the detection task. Confusion matrices and Cohen's kappa scores were used for evaluating classification performance. Two radiologists compared diagnostic performance with the trained model using the external validation set. RESULTS: In the detection task, the model achieved accuracies of 86.36% and 85.37% in the internal and external validation sets, respectively. In the DL model, radiologist 1 and radiologist 2 achieved Cohen's kappa scores of 0.8187, 0.7927, and 0.9077 for four-way classification in the external validation set, respectively. The YOLO DL model illustrated a significantly higher accuracy for intermediate bone tumor classification than radiologist 1 (95.73% vs 88.08%, p = 0.004). CONCLUSIONS: The developed YOLO DL model could be used to assist radiologists at all stages of bone lesion detection and classification in full-field bone radiographs. KEY POINTS: • YOLO DL model can automatically detect bone neoplasms from full-field radiographs in one shot and then simultaneously classify radiographs into normal, benign, intermediate, or malignant. • The dataset used in this retrospective study includes normal bone radiographs. • YOLO can detect even some challenging cases with small volumes.