Osteosarcoma After Total Knee Arthroplasty: A Case Report.

CASE: A 79-year-old woman presented with a periprosthetic fracture 8 years after a total knee arthroplasty (TKA). Radiographs demonstrated tibial implant loosening with severe osteolysis. A high-grade osteosarcoma around the prosthesis was diagnosed, and a supracondylar femoral amputation was performed. After 2 years, no complications have occurred. CONCLUSIONS: A malignant tumor around a TKA is extremely rare. Surgeons should remain vigilant with patients who present with rapidly progressive or very aggressive implant loosening with osteolysis. Owing to its complexity and potentially devastating prognosis, treatment should be guided by a specialist multidisciplinary team. Complex limb salvage procedures or amputation is usually required.

Chondrosarcoma evaluation using hematein-based x-ray staining and high-resolution 3D micro-CT: a feasibility study.

BACKGROUND: Chondrosarcomas are rare malignant bone tumors diagnosed by analyzing radiological images and histology of tissue biopsies and evaluating features such as matrix calcification, cortical destruction, trabecular penetration, and tumor cell entrapment. METHODS: We retrospectively analyzed 16 cartilaginous tumor tissue samples from three patients (51-, 54-, and 70-year-old) diagnosed with a dedifferentiated chondrosarcoma at the femur, a moderately differentiated chondrosarcoma in the pelvis, and a predominantly moderately differentiated chondrosarcoma at the scapula, respectively. We combined a hematein-based x-ray staining with high-resolution three-dimensional (3D) microscopic x-ray computed tomography (micro-CT) for nondestructive 3D tumor assessment and tumor margin evaluation. RESULTS: We detected trabecular entrapment on 3D micro-CT images and followed bone destruction throughout the volume. In addition to staining cell nuclei, hematein-based staining also improved the visualization of the tumor matrix, allowing for the distinction between the tumor and the bone marrow cavity. The hematein-based staining did not interfere with further conventional histology. There was a 5.97 ± 7.17% difference between the relative tumor area measured using micro-CT and histopathology (p = 0.806) (Pearson correlation coefficient r = 0.92, p = 0.009). Signal intensity in the tumor matrix (4.85 ± 2.94) was significantly higher in the stained samples compared to the unstained counterparts (1.92 ± 0.11, p = 0.002). CONCLUSIONS: Using nondestructive 3D micro-CT, the simultaneous visualization of radiological and histopathological features is feasible. RELEVANCE STATEMENT: 3D micro-CT data supports modern radiological and histopathological investigations of human bone tumor specimens. It has the potential for being an integrative part of clinical preoperative diagnostics. KEY POINTS: • Matrix calcifications are a relevant diagnostic feature of bone tumors. • Micro-CT detects all clinically diagnostic relevant features of x-ray-stained chondrosarcoma. • Micro-CT has the potential to be an integrative part of clinical diagnostics.

Chondromyxoid fibroma: A retrospective evaluation of 31 cases.

OBJECTIVES: This study aimed to review a 35-year experience with chondromyxoid fibroma at our institution. PATIENTS AND METHODS: The study retrospectively analyzed the records of 31 consecutive patients (17 males, 14 females; mean age: 30.5±15.7 years; range, 6 to 63 years) with chondromyxoid fibroma who were treated between January 1988 and December 2021. The clinical and radiological characteristics of lesions, tumor volume, and recurrence rates were assessed using the tumor archive of the hospital. RESULTS: The mean follow-up duration was 65.9±42.0 months. Pelvis, proximal tibia, and distal femur were the most common sites of localization. The initial surgical treatment was performed on 27 patients at our clinic, while four patients were referred to the clinic after recurrence. The overall recurrence rate was 16.1%. Intralesional curettage was applied to 21 (77.8%) out of 27 patients. The cavity created after curettage was filled with bone graft (autograft or allograft) in 15 (55.5%) cases. Bone cement was applied in four (14.8%) cases. Resection was applied to five (18.5%) patients. In two (7.4%) cases, intralesional curettage alone was performed. One of these two patients experienced recurrence, resulting in a recurrence rate of 50% in this patient group. No recurrence was observed in other treatment groups. CONCLUSION: Intralesional curettage and filling the defect with bone graft or cement were effective for local control in most cases. Curettage alone was associated with high recurrence rates.

[Osteosarcoma Secondary to Polyostotoic Fibrous Dysplasia of the Ribs].

Sarcomatous transformation of fibrous dysplasia is extremely rare. We present the case of a 54-yearold man with multiple rib masses, multiple enlarged lymph nodes throughout the body, and multiple osteolytic lesions on computed tomography( CT). A positron emission tomography( PET) scan showed abnormal enhancement in each. A needle biopsy of the right supraclavicular fossa lymph node revealed sarcoidosis. Considering the possibility of malignancy associated with sarcoidosis, a rib tumor resection and mediastinal lymph node biopsy were performed to confirm the diagnosis of the rib lesion. The pathology results showed that the rib mass was a low-grade central osteosarcoma and the mediastinal lymph node was sarcoidosis. The distribution of the lesions was consistent with osteosarcoma secondary to multiple fibrous bone dysplasia. As the osteosarcoma was low grade, the patient was followed up. Three years after surgery, there was no increase in residual disease.

Neoplastic and Non-neoplastic Bone Lesions of the Knee.

Numerous anatomical variants are described around the knee, many of which look like bony lesions, so it is important to know them to avoid unnecessary complementary tests and inadequate management. Likewise, several alterations in relation to normal development can also simulate bone lesions.However, numerous pathologic processes frequently affect the knee, including traumatic, inflammatory, infectious, and tumor pathology. Many of these entities show typical radiologic features that facilitate their diagnosis. In other cases, a correct differential diagnosis is necessary for proper clinical management.Despite the availability of increasingly advanced imaging techniques, plain radiography is still the technique of choice in the initial study of many of these pathologies. This article reviews the radiologic characteristics of tumor and nontumor lesions that may appear around the knee to make a correct diagnosis and avoid unnecessary complementary radiologic examinations and inadequate clinical management.

Subungual Osteochondroma of the Great Toe: A Case Report.

Bony outgrowths of the distal phalanx of the great toe have been described in the literature but rarely. These subungual bony outgrowths can be caused by subungual exostosis or subungual osteochondromas. Both of these abnormalities are bony outgrowths with differences in the cartilage cap wherein the exostoses have fibrocartilage, and osteochondromas have hyaline cartilage. The subungual exostosis and osteochondroma that are protruding present symptoms of pain, redness, and deformed nail bed, whereas the nonprotruding osteochondromas have only a lump as the presenting symptom. In both conditions, excision of the lesion and curettage of the base helps prevent a recurrence. Curettage at the end of the excision of the bony outgrowth is required to avoid recurrence. After excision, the specimen should be sent for histopathologic examination to differentiate between the exostosis and osteochondromas, which are underreported in subungual locations, and to rule out malignant transformation. We present a 13-year-old girl with an isolated subungual nonprotruding exostosis of the great toe that was treated by excisional biopsy. The histopathologic examination confirmed it as osteochondroma, which is underreported.

Diagnostic accuracy of contrast-enhanced computed tomography in assessing bone invasion in patients with oral squamous cell carcinoma.

OBJECTIVES: This study aimed to evaluate the diagnostic accuracy of contrast-enhanced computed tomography (CT) in detecting bone invasion in oral squamous cell carcinoma (OSCC) patients and to explore clinicopathological factors associated with its reliability. MATERIALS AND METHODS: 417 patients underwent preoperative contrast-enhanced CT followed by radical surgery. The presence or absence of bone invasion served as the outcome variable, with histopathologic examination of the resection specimen considered the gold standard. Statistical analyses, comprising correlation analyses and the determination of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), were conducted. RESULTS: CT exhibited 76.85% sensitivity, 82.20% specificity, 47.14% PPV, and 89.67% NPV. False-positive and false-negative rates were 11.27% and 5.99%, respectively. Artifacts affected assessment in 44 patients, but not in those with bone invasion. Tumor size, depth of invasion (DOI), tumor localization at the upper jaw, lymphatic invasion, and perineural invasion correlated with incorrect identification of bone invasion (Chi-square, p < 0.05). CONCLUSIONS: Despite utilizing thin-section CT, notable false-positive and false-negative results persisted. Patients with T3 tumors, DOI ≥ 10 mm, or upper jaw tumors are at higher risk for misidentification of bone invasion. Combining multiple methods may enhance diagnostic accuracy, and the integration of artificial intelligence or tracking electrolyte disturbances by tumor depth profiling shows promise for further assessment of bone invasion before histopathology. CLINICAL RELEVANCE: Surgeons should consider these insights when planning tumor resection. Supplementary imaging may be warranted in cases with high risk factors for misidentification. Further methodological advancements are crucial for enhancing diagnostic precision.

Three-dimensional printed custom-made modular talus prosthesis in patients with talus malignant tumor resection.

BACKGROUND: Talar malignant tumor is extremely rare. Currently, there are several alternative management options for talus malignant tumor including below-knee amputation, tibio-calcaneal arthrodesis, and homogenous bone transplant while their shortcomings limited the clinical application. Three-dimensional (3D) printed total talus prosthesis in talus lesion was reported as a useful method to reconstruct talus, however, most researches are case reports and its clinical effect remains unclear. Therefore, the current study was to explore the application of 3D printed custom-made modular prosthesis in talus malignant tumor. METHODS: We retrospectively analyzed the patients who received the 3D printed custom-made modular prosthesis treatment due to talus malignant tumor in our hospital from February 2016 to December 2021. The patient's clinical data such as oncology outcome, operation time, and volume of blood loss were recorded. The limb function was evaluated with the Musculoskeletal Tumor Society 93 (MSTS-93) score, The American Orthopedic Foot and Ankle Society (AOFAS) score; the ankle joint ranges of motion as well as the leg length discrepancy were evaluated. Plain radiography and Tomosynthesis-Shimadzu Metal Artefact Reduction Technology (T-SMART) were used to evaluate the position of prosthesis and the osseointegration. Postoperative complications were recorded. RESULTS: The average patients' age and the follow-up period were respectively 31.5 ± 13.1 years; and 54.8 months (range 26-72). The medium operation time was 2.4 ± 0.5 h; the intraoperative blood loss was 131.7 ± 121.4 ml. The mean MSTS-93 and AOFAS score was 26.8 and 88.5 respectively. The average plantar flexion, dorsiflexion, varus, and valgus were 32.5, 9.2, 10.8, and 5.8 degree respectively. One patient had delayed postoperative wound healing. There was no leg length discrepancy observed in any patient and good osseointegration was observed on the interface between the bone and talus prosthesis in all subjects. CONCLUSION: The modular structure of the prosthesis developed in this study seems to be convenient for prosthesis implantation and screws distribution. And the combination of solid and porous structure improves the initial stability and promotes bone integration. Therefore, 3D printed custom-made modular talus prosthesis could be an alternative option for talus reconstruction in talus malignant tumor patients.

3D-Printed custom-made hemipelvic endoprosthetic reconstruction following periacetabular tumor resection: utilizing a novel classification system.

BACKGROUND: Customized 3D-printed pelvic implants with a porous structure have revolutionized periacetabular pelvic defect reconstruction after tumor resection, offering improved osteointegration, long-term stability, and anatomical fit. However, the lack of an established classification system hampers implementation and progress. METHODS: We formulated a novel classification system based on pelvic defect morphology and 3D-printed hemipelvis endoprostheses. It integrates surgical approach, osteotomy guide plate and prosthesis design, postoperative rehabilitation plans, and perioperative processes. RESULTS: Retrospectively analyzing 60 patients (31 males, 29 females), we classified them into Type A (15 patients: Aa = 6, Ab = 9), Type B (27 patients: Ba = 15, Bb = 12), Type C (17 patients). All underwent customized osteotomy guide plate-assisted tumor resection and 3D-printed hemipelvic endoprosthesis reconstruction. Follow-up duration was median 36.5 ± 15.0 months (range, 6 to 74 months). The mean operating time was 430.0 ± 106.7 min, intraoperative blood loss 2018.3 ± 1305.6 ml, transfusion volume 2510.0 ± 1778.1 ml. Complications occurred in 13 patients (21.7%), including poor wound healing (10.0%), deep prosthesis infection (6.7%), hip dislocation (3.3%), screw fracture (1.7%), and interface loosening (1.7%). VAS score improved from 5.5 ± 1.4 to 1.7 ± 1.3, MSTS-93 score from 14.8 ± 2.5 to 23.0 ± 5.6. Implant osseointegration success rate was 98.5% (128/130), with one Type Ba patient experiencing distal prosthesis loosening. CONCLUSION: The West China classification may supplement the Enneking and Dunham classification, enhancing interdisciplinary communication and surgical outcomes. However, further validation and wider adoption are required to confirm clinical effectiveness.

Can FDG-PET assess the response to chemotherapy and predict tissue necrosis in osteosarcoma and Ewing sarcoma?

INTRODUCTION: Osteosarcoma (OS) and Ewing sarcoma (ES) represent the pediatric population's most common malignant bone tumors. 18-Fluorodeoxyglucose positron emission tomography has been shown to be effective in both the diagnostic and staging phases of cancer treatment. In recent years, some studies have also explored the possibility that FDG-PET could have a prognostic role.

Biomimetic design and clinical application of Ti-6Al-4V lattice hemipelvis prosthesis for pelvic reconstruction.

OBJECTIVE: This study aims to biomimetic design a new 3D-printed lattice hemipelvis prosthesis and evaluate its clinical efficiency for pelvic reconstruction following tumor resection, focusing on feasibility, osseointegration, and patient outcomes. METHODS: From May 2020 to October 2021, twelve patients with pelvic tumors underwent tumor resection and subsequently received 3D-printed lattice hemipelvis prostheses for pelvic reconstruction. The prosthesis was strategically incorporated with lattice structures and solid to optimize mechanical performance and osseointegration. The pore size and porosity were analyzed. Patient outcomes were assessed through a combination of clinical and radiological evaluations. RESULTS: Multiple pore sizes were observed in irregular porous structures, with a wide distribution range (approximately 300-900 µm). The average follow-up of 34.7 months, ranging 26 from to 43 months. One patient with Ewing sarcoma died of pulmonary metastasis 33 months after surgery while others were alive at the last follow-up. Postoperative radiographs showed that the prosthesis's position was consistent with the preoperative planning. T-SMART images showed that the host bone was in close and tight contact with the prosthesis with no gaps at the interface. The average MSTS score was 21 at the last follow-up, ranging from 18 to 24. There was no complication requiring revision surgery or removal of the 3D-printed hemipelvis prosthesis, such as infection, screw breakage, and prosthesis loosening. CONCLUSION: The newly designed 3D-printed lattice hemipelvis prosthesis created multiple pore sizes with a wide distribution range and resulted in good osteointegration and favorable limb function.

A CT-based radiomics nomogram for predicting histologic grade and outcome in chondrosarcoma.

OBJECTIVE: The preoperative identification of tumor grade in chondrosarcoma (CS) is crucial for devising effective treatment strategies and predicting outcomes. The study aims to build and validate a CT-based radiomics nomogram (RN) for the preoperative identification of tumor grade in CS, and to evaluate the correlation between the RN-predicted tumor grade and postoperative outcome. METHODS: A total of 196 patients (139 in the training cohort and 57 in the external validation cohort) were derived from three different centers. A clinical model, radiomics signature (RS) and RN (which combines significant clinical factors and RS) were developed and validated to assess their ability to distinguish low-grade from high-grade CS with area under the curve (AUC). Additionally, Kaplan-Meier survival analysis was applied to examine the association between RN-predicted tumor grade and recurrence-free survival (RFS) of CS. The predictive accuracy of the RN was evaluated using Harrell's concordance index (C-index), hazard ratio (HR) and AUC. RESULTS: Size, endosteal scalloping and active periostitis were selected to build the clinical model. Three radiomics features, based on CT images, were selected to construct the RS. Both the RN (AUC, 0.842) and RS (AUC, 0.835) were superior to the clinical model (AUC, 0.776) in the validation set (P = 0.003, 0.040, respectively). A correlation between Nomogram score (Nomo-score, derived from RN) and RFS was observed through Kaplan-Meier survival analysis in the training and test cohorts (log-rank P < 0.050). Patients with high Nomo-score tumors were 2.669 times more likely to suffer recurrence than those with low Nomo-score tumors (HR, 2.669, P < 0.001). CONCLUSIONS: The CT-based RN performed well in predicting both the histologic grade and outcome of CS.

Advancing musculoskeletal tumor diagnosis: Automated segmentation and predictive classification using deep learning and radiomics.

OBJECTIVES: Musculoskeletal (MSK) tumors, given their high mortality rate and heterogeneity, necessitate precise examination and diagnosis to guide clinical treatment effectively. Magnetic resonance imaging (MRI) is pivotal in detecting MSK tumors, as it offers exceptional image contrast between bone and soft tissue. This study aims to enhance the speed of detection and the diagnostic accuracy of MSK tumors through automated segmentation and grading utilizing MRI. MATERIALS AND METHODS: The research included 170 patients (mean age, 58 years ±12 (standard deviation), 84 men) with MSK lesions, who underwent MRI scans from April 2021 to May 2023. We proposed a deep learning (DL) segmentation model MSAPN based on multi-scale attention and pixel-level reconstruction, and compared it with existing algorithms. Using MSAPN-segmented lesions to extract their radiomic features for the benign and malignant classification of tumors. RESULTS: Compared to the most advanced segmentation algorithms, MSAPN demonstrates better performance. The Dice similarity coefficients (DSC) are 0.871 and 0.815 in the testing set and independent validation set, respectively. The radiomics model for classifying benign and malignant lesions achieves an accuracy of 0.890. Moreover, there is no statistically significant difference between the radiomics model based on manual segmentation and MSAPN segmentation. CONCLUSION: This research contributes to the advancement of MSK tumor diagnosis through automated segmentation and predictive classification. The integration of DL algorithms and radiomics shows promising results, and the visualization analysis of feature maps enhances clinical interpretability.

Osteoid osteoma of the distal radius presenting as an inconspicuous swelling in a young child.

Osteoid osteoma is a benign osteoblastic tumour with a predilection for the lower extremity that rarely affects the forearm. It is commonly seen in adolescents and young adults, and is seldom diagnosed in the paediatric age group. We report a boy in his early childhood who presented with a swelling over the distal forearm, which was incidentally noted by the mother 3 months ago. Plain radiographs showed diffuse sclerosis of the dorsal cortex of the distal radius. CT scan showed a central lucent nidus in the intramedullary region and surrounding sclerosis in the radial metaphysis, confirming the diagnosis of osteoid osteoma. The patient was successfully treated by surgical en bloc resection of the nidus and was asymptomatic at 1-year follow-up. Non-specific symptoms at presentation make it a challenge to diagnose osteoid osteoma in children and it needs to be considered in the differential diagnosis when radiographs show lytic lesions in the bone.

DECIDE: A decoupled semantic and boundary learning network for precise osteosarcoma segmentation by integrating multi-modality MRI.

Automated Osteosarcoma Segmentation in Multi-modality MRI (AOSMM) holds clinical significance for effective tumor evaluation and treatment planning. However, the precision of AOSMM is challenged by the diverse characteristics of multi-modality MRI and the inherent heterogeneity and boundary ambiguity of osteosarcoma. While numerous methods have made significant strides in automated osteosarcoma segmentation, they primarily focused on the use of a single MRI modality and overlooked the potential benefits of integrating complementary information from other MRI modalities. Furthermore, they did not adequately model the long-range dependencies of complex tumor features, which may lead to insufficiently discriminative feature representations. To this end, we propose a decoupled semantic and boundary learning network (DECIDE) to achieve precise AOSMM with three functional modules. The Multi-modality Feature Fusion and Recalibration (MFR) module adaptively fuses and recalibrates multi-modality features by exploiting their channel-wise dependencies to compute low-rank attention weights for effectively aggregating useful information from different MRI modalities, which promotes complementary learning between multi-modality MRI and enables a more comprehensive tumor characterization. The Lesion Attention Enhancement (LAE) module employs spatial and channel attention mechanisms to capture global contextual dependencies over local features, significantly enhancing the discriminability and representational capacity of intricate tumor features. The Boundary Context Aggregation (BCA) module further enhances semantic representations by utilizing boundary information for effective context aggregation while also ensuring intra-class consistency in cases of boundary ambiguity. Substantial experiments demonstrate that DECIDE achieves exceptional performance in osteosarcoma segmentation, surpassing state-of-the-art methods in terms of accuracy and stability.

Osteosarcoma-targeted Cu and Ce based oxide nanoplatform for NIR II fluorescence/magnetic resonance dual-mode imaging and ros cascade amplification along with immunotherapy.

BACKGROUND: As the lethal bone tumor, osteosarcoma often frequently occurs in children and adolescents with locally destructive and high metastasis. Distinctive kinds of nanoplatform with high therapeutical effect and precise diagnosis for osteosarcoma are urgently required. Multimodal optical imaging and programmed treatment, including synergistic photothermal-chemodynamic therapy (PTT-CDT) elicits immunogenetic cell death (ICD) is a promising strategy that possesses high bio-imaging sensitivity for accurate osteosarcoma delineating as well as appreciable therapeutic efficacy with ignorable side-effects. METHODS AND RESULTS: In this study, mesoporous Cu and Ce based oxide nanoplatform with Arg-Gly-Asp (RGD) anchoring is designed and successfully constructed. After loading with indocyanine green, this nanoplatform can be utilized for precisely targeting and efficaciously ablating against osteosarcoma via PTT boosted CDT and the closely following ICD stimulation both in vitro and in vivo. Besides, it provides off-peak fluorescence bio-imaging in the second window of near-infrared region (NIR II, 1000-1700 nm) and Magnetic resonance signal, serves as the dual-mode contrast agents for osteosarcoma tissue discrimination. CONCLUSION: Tumor targeted Cu&Ce based mesoporous nanoplatform permits efficient osteosarcoma suppression and dual-mode bio-imaging that opens new possibility for effectively diagnosing and inhibiting the clinical malignant osteosarcoma.

Osteochondrolipoma of the foot treated by surgical excision: a case report and literature review.

BACKGROUND: Osteochondromas, classified as a new benign subtype of lipomas and characterised by chondroid and osseous differentiation, are rare lesions that have been infrequently reported in previous literature. The maxillofacial region was reported as the most frequent localization, with infrequent occurrence in the lower limb. This paper represents the first documented case report of osteochondrolipoma in the foot. CASE PRESENTATION: A 51-year-old male patient presented with a chief complaint of right foot pain at the plantar aspect, accompanied by the observation of swelling between the first and the second metatarsal shafts. His complaint of pain and swelling started 10 and 4 years prior, respectively. Since their onset, both symptoms have progressed in nature. Imaging revealved a large mass exhibiting a nonhomogenous composition of fibrous tissue and bony structures. Surgical intervention through total excision was indicated. CONCLUSION: Osteochodrolipoma is a benign lesion that can affect the foot leading to decreased functionality of the foot due to the pain and swelling. Surgical excision is the recommended approach for this lesion, providing both symptomatic relief and confirmation of the diagnosis through histopathological examination.

Desmoplastic fibroma in a child: a 9-year follow-up case report.

BACKGROUND: Desmoplastic fibroma is an extremely rare primary bone tumor. Its characteristic features include bone destruction accompanied by the formation of soft tissue masses. This condition predominantly affects individuals under the age of 30. Since its histology is similar to desmoid-type fibromatosis, an accurate diagnosis before operation is difficult. Desmoplastic fibroma is resistant to chemotherapy, and the efficacy of radiotherapy is uncertain. Surgical excision is preferred for treatment, but it entails high recurrence. Further, skeletal reconstruction post-surgery is challenging, especially in pediatric cases. CASE PRESENTATION: Nine years ago, a 14-year-old male patient presented with a 4-year history of progressive pain in his left wrist. Initially diagnosed as fibrous dysplasia by needle biopsy, the patient underwent tumor resection followed by free vascularized fibular proximal epiphyseal transfer for wrist reconstruction. However, a histological examination confirmed a diagnosis of desmoplastic fibroma. The patient achieved bone union and experienced a recurrence in the ipsilateral ulna 5 years later, accompanied by a wrist deformity. He underwent a second tumor resection and wrist arthrodesis in a single stage. The most recent annual follow-up was in September 2023; the patient had no recurrence and was satisfied with the surgery. CONCLUSIONS: Desmoplastic fibroma is difficult to diagnose and treat, and reconstruction surgery after tumor resection is challenging. Close follow-up by experienced surgeons may be beneficial for prognosis.

Automated Detection and Segmentation of Bone Metastases on Spine MRI Using U-Net: A Multicenter Study.

OBJECTIVE: To develop and evaluate a deep learning model for automated segmentation and detection of bone metastasis on spinal MRI. MATERIALS AND METHODS: We included whole spine MRI scans of adult patients with bone metastasis: 662 MRI series from 302 patients (63.5 ± 11.5 years; male:female, 151:151) from three study centers obtained between January 2015 and August 2021 for training and internal testing (random split into 536 and 126 series, respectively) and 49 MRI series from 20 patients (65.9 ± 11.5 years; male:female, 11:9) from another center obtained between January 2018 and August 2020 for external testing. Three sagittal MRI sequences, including non-contrast T1-weighted image (T1), contrast-enhanced T1-weighted Dixon fat-only image (FO), and contrast-enhanced fat-suppressed T1-weighted image (CE), were used. Seven models trained using the 2D and 3D U-Nets were developed with different combinations (T1, FO, CE, T1 + FO, T1 + CE, FO + CE, and T1 + FO + CE). The segmentation performance was evaluated using Dice coefficient, pixel-wise recall, and pixel-wise precision. The detection performance was analyzed using per-lesion sensitivity and a free-response receiver operating characteristic curve. The performance of the model was compared with that of five radiologists using the external test set. RESULTS: The 2D U-Net T1 + CE model exhibited superior segmentation performance in the external test compared to the other models, with a Dice coefficient of 0.699 and pixel-wise recall of 0.653. The T1 + CE model achieved per-lesion sensitivities of 0.828 (497/600) and 0.857 (150/175) for metastases in the internal and external tests, respectively. The radiologists demonstrated a mean per-lesion sensitivity of 0.746 and a mean per-lesion positive predictive value of 0.701 in the external test. CONCLUSION: The deep learning models proposed for automated segmentation and detection of bone metastases on spinal MRI demonstrated high diagnostic performance.