A Deep Learning Model Enhances Clinicians' Diagnostic Accuracy to More Than 96% for Anterior Cruciate Ligament Ruptures on Magnetic Resonance Imaging.

PURPOSE: To develop a deep learning model to accurately detect anterior cruciate ligament (ACL) ruptures on magnetic resonance imaging (MRI) and to evaluate its effect on the diagnostic accuracy and efficiency of clinicians. METHODS: A training dataset was built from MRIs acquired from January 2017 to June 2021, including patients with knee symptoms, irrespective of ACL ruptures. An external validation dataset was built from MRIs acquired from January 2021 to June 2022, including patients who underwent knee arthroscopy or arthroplasty. Patients with fractures or prior knee surgeries were excluded in both datasets. Subsequently, a deep learning model was developed and validated using these datasets. Clinicians of varying expertise levels in sports medicine and radiology were recruited, and their capacities in diagnosing ACL injuries in terms of accuracy and diagnosing time were evaluated both with and without artificial intelligence (AI) assistance. RESULTS: A deep learning model was developed based on the training dataset of 22,767 MRIs from 5 centers and verified with external validation dataset of 4,086 MRIs from 6 centers. The model achieved an area under the receiver operating characteristic curve of 0.987 and a sensitivity and specificity of 95.1%. Thirty-eight clinicians from 25 centers were recruited to diagnose 3,800 MRIs. The AI assistance significantly improved the accuracy of all clinicians, exceeding 96%. Additionally, a notable reduction in diagnostic time was observed. The most significant improvements in accuracy and time efficiency were observed in the trainee groups, suggesting that AI support is particularly beneficial for clinicians with moderately limited diagnostic expertise. CONCLUSIONS: This deep learning model demonstrated expert-level diagnostic performance for ACL ruptures, serving as a valuable tool to assist clinicians of various specialties and experience levels in making accurate and efficient diagnoses. LEVEL OF EVIDENCE: Level III, retrospective comparative case series.

Application of Magnetic Resonance Imaging in Neoadjuvant Treatment of Pancreatic Ductal Adenocarcinoma.

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest malignant tumors of the human digestive system. Due to its insidious onset, many patients have already lost the opportunity for radical resection upon tumor diagnosis. In recent years, neoadjuvant treatment for patients with borderline resectable PDAC has been recommended by multiple guidelines to increase the resection rate of radical surgery and improve the postoperative survival. However, further developments are required to accurately assess the tumor response to neoadjuvant therapy and to select the population suitable for such treatment. Reductions in drug toxicity and the number of neoadjuvant cycles are also critical. At present, the clinical evaluation of neoadjuvant treatment is mainly based on several serological and imaging indicators; however, the unique characteristics of PDAC and the insufficient sensitivity and specificity of the markers render this system ineffective. The imaging evaluation system, magnetic resonance imaging (MRI), has its own unique imaging advantages compared with computed tomography (CT) and other imaging examinations. One key advantage is the ability to reflect the changes more rapidly in tumor tissue components, such as the degree of fibrosis, microvessel density, and tissue hypoxia. It can also perform multiparameter quantitative analysis of tumor tissue and changes, attributing to its increasingly important role in imaging evaluation, and potentially the evaluation of neoadjuvant treatment of pancreatic cancer, as several current articles have studied. At the same time, owing to the complexity of MRI and some of its limitations, its wider application is limited. Compared with CT imaging, few relevant studies have been conducted. In this review article, we will investigate and summarize the advantages, limitations, and future development of MRI in the evaluation of neoadjuvant treatment of PDAC. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Utility of Advanced DWI in the Detection of Spinal Cord Microstructural Alterations and Assessment of Neurologic Function in Cervical Spondylotic Myelopathy Patients.

BACKGROUND: Diffusion-weighted imaging (DWI) can quantify the microstructural changes in the spinal cord. It might be a substitute for T2 increased signal intensity (ISI) for cervical spondylotic myelopathy (CSM) evaluation and prognosis. PURPOSE: The purpose of the study is to investigate the relationship between DWI metrics and neurologic function of patients with CSM. STUDY TYPE: Retrospective. POPULATION: Forty-eight patients with CSM (18.8% females) and 36 healthy controls (HCs, 25.0% females). FIELD STRENGTH/SEQUENCE: 3 T; spin-echo echo-planar imaging-DWI; turbo spin-echo T1/T2; multi-echo gradient echo T2*. ASSESSMENT: For patients, conventional MRI indicators (presence and grades of T2 ISI), DWI indicators (neurite orientation dispersion and density imaging [NODDI]-derived isotropic volume fraction [ISOVF], intracellular volume fraction, and orientation dispersion index [ODI], diffusion tensor imaging [DTI]-derived fractional anisotropy [FA] and mean diffusivity [MD], and diffusion kurtosis imaging [DKI]-derived FA, MD, and mean kurtosis), clinical conditions, and modified Japanese Orthopaedic Association (mJOA) were recorded before the surgery. Neurologic function improvement was measured by the 3-month follow-up recovery rate (RR). For HCs, DWI, and mJOA were measured as baseline comparison. STATISTICAL TESTS: Continuous (categorical) variables were compared between patients and HCs using Student's t-tests or Mann-Whitney U tests (chi-square or Fisher exact tests). The relationships between DWI metrics/conventional MRI findings, and the pre-operative mJOA/RR were assessed using correlation and multivariate analysis. P < 0.05 was considered statistically significant. RESULTS: Among patients, grades of T2 ISI were not correlated with pre-surgical mJOA/RR (P = 0.717  and 0.175, respectively). NODDI ODI correlated with pre-operative mJOA (r = -0.31). DTI FA, DKI FA, and NODDI ISOVF were correlated with the recovery rate (r = 0.31, 0.41, and -0.34, respectively). In multivariate analysis, NODDI ODI (DTI FA, DKI FA, NODDI ISOVF) significantly contributed to the pre-operative mJOA (RR) after adjusting for age. DATA CONCLUSION: DTI FA, DKI FA, and NODDI ISOVF are predictors for prognosis in patients with CSM. NODDI ODI can be used to evaluate CSM severity. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 5.

Predicting postoperative recovery in cervical spondylotic myelopathy: construction and interpretation of T2*-weighted radiomic-based extra trees models.

OBJECTIVES: Conventional MRI may not be ideal for predicting cervical spondylotic myelopathy (CSM) prognosis. In this study, we used radiomics in predicting postoperative recovery in CSM. We aimed to develop and validate radiomic feature-based extra trees models. METHODS: There were 151 patients with CSM who underwent preoperative T2-/ T2*-weighted imaging (WI) and surgery. They were divided into good/poor outcome groups based on the recovery rate. Datasets from multiple scanners were randomised into training and internal validation sets, while the dataset from an independent scanner was used for external validation. Radiomic features were extracted from the transverse spinal cord at the maximum compressed level. Threshold selection algorithm, collinearity removal, and tree-based feature selection were applied sequentially in the training set to obtain the optimal radiomic features. The classification of intramedullary increased signal on T2/T2*WI and compression ratio of the spinal cord on T2*WI were selected as the conventional MRI features. Clinical features were age, preoperative mJOA, and symptom duration. Four models were constructed: radiological, radiomic, clinical-radiological, and clinical-radiomic. An AUC significantly > 0.5 was considered meaningful predictive performance based on the DeLong test. The mean decrease in impurity was used to measure feature importance. p < 0.05 was considered statistically significant. RESULTS: On internal and external validations, AUCs of the radiomic and clinical-radiomic models, and radiological and clinical-radiological models ranged from 0.71 to 0.81 (significantly > 0.5) and 0.40 to 0.55, respectively. Wavelet-LL first-order variance was the most important feature in the radiomic model. CONCLUSION: Radiomic features, especially wavelet-LL first-order variance, contribute to meaningful predictive models for CSM prognosis. KEY POINTS: • Conventional MRI features may not be ideal in predicting prognosis. • Radiomics provides greater predictive efficiency in the recovery from cervical spondylotic myelopathy.

[Applications of Artificial Intelligence in Musculoskeletal System Imaging].

Artificial intelligence (AI) represents the latest wave of computer revolution and is considered revolutionary technology in many industries including healthcare. AI has been applied in medical imaging mainly due to the improvement of computational learning,big data mining,and innovations of neural network architecture. AI can improve the efficiency and accuracy of imaging diagnosis and reduce medical cost;also,it can be used to predict the disease risk. In this article we summarize and analyze the application of AI in musculoskeletal imaging.

Intraosseous schwannoma of the mobile spine: a report of twenty cases.

PURPOSE: To clarify the clinical features, surgical strategies, and outcomes of intraosseous schwannoma (IOS) of the mobile spine. METHODS: We retrospectively reviewed patients with primary benign spinal schwannoma who underwent surgery in our orthopedic department. RESULTS: A total of 101 patients with primary benign schwannoma located in the mobile spine underwent surgery in our orthopedic department from 2005 to 2015. Twenty-five patients presented with aggressive features. Twenty patients were regularly followed up, twelve with lesions in the cervical spine, six with lesions in the thoracic region, and two with lesions in the lumbar spine. Preoperative CT-guided biopsy was performed in fourteen cases; the accuracy of diagnosis was 100%, and IOS is not histologically different from conventional schwannoma. The computed tomography (CT) scan revealed expansile and osteolytic bone destruction in all these cases, with six patients having pathological fracture. On T2-weighted magnetic resonance imaging, the lobulated schwannomas showed heterogeneous signal intensity and significant heterogeneous enhancement on post-contrast images. Gross total resection was performed in seventeen patients and subtotal resection in three. Tumor-involved nerve roots resection were documented to decrease local recurrence in fourteen cases. The visual analog scale score decreased from 5.66 ± 1.79 preoperatively to 1.16 ± 1.77 at the final follow-up. No local recurrence was noticed at the final follow-up. CONCLUSION: CT-guided biopsy is effective for the preoperative diagnosis of spinal IOS. Total resection is the optimal treatment for IOS, whereas subtotal resection could be an alternative choice for high-risk cases. These slides can be retrieved under electronic supplementary material.

[Value of Diffusion-weighted Imaging in Evaluating the Activity of Sacroiliitis in Ankylosing Spondylitis].

Objective To investigate the clinical value of diffusion-weighted imaging (DWI) for evaluating the activity of sacroiliitis in ankylosing spondylitis (AS).Methods Totally 73 AS patients were prospectively enrolled and divided into active group (n=43) and chronic group (n=30) according to Bath ankylosing spondylitis disease activity index (BASDAI) scores and laboratory findings. Conventional magnetic resonance imaging (MRI) and DWI were performed in all subjects. Apparent diffusion coefficient (ADC) values of subchondral lesions in sacroiliac joint were independently measured by two radiologists,and the relative ADC (rADC) values were calculated. ADC and rADC values were compared between active and chronic groups. The efficiencies of ADC and rADC values for differentiating the activity of sacroiliitis were analyzed. In addition,the correlation coefficients of ADC values,rADC values,and BASDAI scores were calculated.Results The ADC and rADC values in the active group were (0.667±0.122)×10 -3 mm 2/s and (1.715±0.343)×10 -3 mm 2/s,respectively,which were significantly higher than those of the chronic group [(0.492±0.0651)×10 -3 mm 2/s and (1.289±0.209)×10 -3 mm 2/s,respectively)] (P<0.0001). The agreement of measurement results between two radiologists was good,and all the interclass correlation coefficients were >0.81. The correlation coefficients of ADC value and rADC value with BASDAI scores were 0.82 and 0.80,respectively (P<0.0001). The optimal cutoff values of ADC value and rADC value for differentiating AS activity were 0.545×10 -3 mm 2/s and 1.467×10 -3 mm 2/s,respectively,The specificity was 81.8% for both indicators,and the sensitivity was 92.0% and 88.0%,respectively.Conclusion DWI is helpful in the quantitative assessment of the activity of sacroiliitis in AS patients.

Percutaneous radiofrequency ablation for spinal osteoid osteoma and osteoblastoma.

PURPOSES: We sought to verify the efficacy and safety of RFA in spinal OO and osteoblastomas (OB) (Enneking Stage 2, S2). METHODS: We retrospectively reviewed patients treated in our hospital. Surgical resection was indicated for Enneking Stage 3 OB. RFA indications for spinal OO and OB (S2) were no neurological deficits, complete bone cortex around the lesion on computed tomography (CT), and cerebrospinal fluid between a lesion and the spinal cord/nerve root on magnetic resonance imaging. Abundant cerebrospinal fluid (more than 1.0 mm) between the lesion and nerve root/spinal cord was preferred to prevent neurological damage by heat. Otherwise, surgery was recommended. The minimum follow-up was 24 months. RESULTS: Ten patients were treated with CT-guided percutaneous RFA, including three with OB and seven with OO. No patients had neurological deficits or scoliosis. In OO patients, the average visual analog scale (VAS) scores were 7.6/10 (range 6-10) before RFA. In OB cases, the VAS scores were 8, 7, and 9 before RFA. Nine patients had a one-stage biopsy and then RFA, and one patient had a two-stage procedure (biopsy before RFA). The average RFA time for OO was 10 min (range 4-12). In the three OB cases, the RFA time was 12, 12, and 24 min. The time of the whole produce was 98 min (range 65-130 min). All 10 patients were followed-up. The average follow-up time of OO was 46.6 months (range 24-66). Six patients were free of pain, except one who suffered occasional pain with VAS 2/10. The three OB cases were free of pain at 24, 26, and 26 months. CONCLUSION: CT-guided percutaneous RFA is a safe and effective treatment for spinal OO and S2 OB, especially in lesions with no neurological deficits and intact cortical bone. Cerebrospinal fluid around the lesion is an appropriate indication for percutaneous RFA.

Preoperative Evaluation of the Grade and Stability of Osteochondritis Dissecans of the Knee by Magnetic Resonance Imaging.

Objective To assess the diagnostic performance of magnetic resonance imaging (MRI) in the grading of osteochondritis dissecans (OCD) of the knee.Methods Totally 47 patients with OCD of the knee confirmed by arthroscopy were retrospectively enrolled in this study.The OCD lesions were classified into four stages according to classification system of the International Cartilage Repair Society.Two radiologists analyzed all MRI findings independently,and the results were compared with those of arthroscopy.Sensitivity,specificity,and accuracy were calculated.Kappa value were calculated to quantify inter-observer agreement of the diagnostic OCD grade between two doctors by MRI.Specificity,sensitivity,and accuracy of MRI criteria indicating instability for detection of OCD instability were calculated.Results Of these 47 patients with 48 OCD lesions,stages Ⅰ,Ⅱ,Ⅲ,and Ⅲ lesions were detected in 4,8,16,and 20 patients,respectively.The specificity,sensitivity,and accuracy for the diagnosis of OCD stability were 75.0% (83.3%),88.9% (86.1%),and 85.4% (85.4%) for observer l (2),and the agreement of OCD grade between these two readers was substantial with a Kappa value of 0.82.The specificity,sensitivity,and accuracy of MRI criteria for the detection of OCD instability including high T2 signal intensity at the interface between the OCD and the underlying bone,multiple cysts or a single cyst of>5 mm in diameter surrounding OCD lesions,high T2 signal intensity cartilage fracture line traversing the articular cartilage,and osteochondral defect were 83.3%,80.6%,and 81.3%;75.0%,72.2%,and 72.9%;66.7%,69.4%,and 68.8%;100%,86.1%,and 89.6%,respectively. Conclusions Osteochondral defect is the most specific MRI sign for diagnosing instable OCD of the knee,whereas osteochondral fracture line has the lowest accuracy.MRI is a useful method to evaluate the grade and stability of OCD of the knee.

Computed Tomography and Magnetic Resonance Manifestations of Spinal Pigmented Villonodular Synovitis.

OBJECTIVE: To study computed tomography (CT) and magnetic resonance imaging (MRI) findings of spinal pigmented villonodular synovitis (PVNS), with the aim of improving diagnostic accuracy. METHODS: Findings from CT (n = 7) and MRI (n = 5) of 7 cases of spinal PVNS, diagnosed by pathological examination, were retrospectively analyzed. RESULTS: One case presented with a small lesion, whereas the remaining 6 cases showed lobulated soft tissue masses centering on the facet joints, with lytic expansive destruction of the bone. The CT density of lesions in 3 cases was similar to that of the surrounding muscles, whereas the remaining 4 cases had lesions that were slightly hypodense relative to muscle. Among 5 cases for which MRI was available, 1 lesion was small with a hypointense edge, 1 manifested homogenous intensity without any obvious hypointense nodule, and multiple hypointensity nodules were visible in the remaining cases. CONCLUSIONS: The CT and MR manifestations of spinal PVNS have certain characteristics and can provide the basis for the preoperative diagnosis.

Tri-planar computed tomographic projection of three segments of the middle turbinate in a Chinese population.

To investigate the exact anatomical characteristics and variations of three segments of middle turbinate bones in a Chinese population using Tri-planar computed tomographic (CT) projections. The axial, sagittal, and coronal thin-slice reconstructed CT images of the middle turbinate were obtained from 200 bones of 100 normal adult Chinese subjects (54 males and 46 females). The middle turbinate bones were divided into three segments: vertical, basal lamella, and horizontal. The anterior vertical segment is a sagittal plane best seen on coronal images. Forty-eight bones of the vertical segment were pneumatized on both sides, 14 bones were pneumatized on one side and 21 bones had paradoxical curvature. The distribution of pneumatized concha bullosa bones (based on shape) was as follows: 41 lamellar, 5 bulbous, and 7 extensive types. The basal lamella of the middle turbinate (an L-shaped bony plate) was easily observed on axial and sagittal imaging. On the axial images, the anterior vertical portion of the bones was classified according to shape: 44 linear, 116 curved, and 40 angular. On the sagittal images, it was apparent that most of the horizontal portion was not at a real horizontal plane but on a slope, with an average angle of 127° in the anterior direction. Racial differences may cause the discrepancy between our results and those of other studies. The present study describes the anatomical characteristics and variations of the middle turbinate in the Chinese in Asia, knowledge of which is essential to avoid complications during surgery.

[Value of multiple b-value diffusion-weighted imaging for differentiation of benign and malignant pulmonary masses].

OBJECTIVE: To explore the value of multiple b-value diffusion-weighted imaging (DWI) for differentiation of benign and malignant pulmonary masses. METHODS: Thirty-eight patients were examined by routine sequences and DWI pulse sequence. DWI was acquired through a single-shot echo-planar imaging combined with a respiratory-triggered mode and parallel acquisition. Nine b values ranging from 0 to 1500 s/mm(2) (0, 50, 100, 150, 200, 400, 600, 1000, 1500 s/mm(2)) were used. The intravoxel incoherent motion model was applied to estimate pure diffusion coefficient D, perfusion-related diffusion coefficient D(*), and perfusion fraction f. Mann-Whitney U test was used to compare all measured parameters between benign and malignant groups. The diagnostic performance of the related parameters was evaluated with receiver operating characteristics (ROC) analysis. RESULTS: Of these 38 patients, 30 were pathologically confirmed and 8 were diagnosed based on clinical data. There were 23 lung malignant masses and 15 benign lesions. A significant reduction of D was found in malignant group than in benign group (Z=3.308, P=0.001), while no significant differences in D(*)(Z=1.646, P=0.100) and f(Z=1.254, P=0.210) were observed between the two groups. The area under the ROC curve for D value (0.839) was largest. When the cutoff value was selected as 0.90×10(-3) mm(2)/s, the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of diagnosing malignant masses were 95.7%, 80.0%, 90.9%, 91.7%, and 88.9%,respectively. CONCLUSION: The D value in multiple b-value DWI has certain significance in differentiating the benign and malignant pulmonary masses and has the best diagnostic efficiency.

Epidemiological survey of ossification of the ligamentum flavum in thoracic spine: CT imaging observation of 993 cases.

OBJECTIVE: To investigate the characteristics of epidemiological distribution of the ossification of the ligamentum flavum (OLF) in the thoracic spine including the incidence, segmental distribution, and shape. METHODS: Chest spiral computed tomography scans of 993 cases (male 506, female 487, mean age 60 years, range 5-102 years) who presented due to chest symptoms were analyzed with axial slices combined with sagittal slices. The conditions of OLF in the thoracic spine, including segments, thickness, location, and dural sac compression, were recorded. Prevalence was standardized according to the "Age Structure of Population in Beijing 2008". RESULTS: Among the population investigated, the standardized prevalence rate was 63.9 %. The standardized prevalence rate for males (68.5 %) was higher than that for females (59.0 %). The highest prevalence rate of OLF was in the 50-59 years age group (79.2 %); however, high density originated it can be found in individuals aged 10-19 years. The comparison of different thoracic segments showed that T10-11 (44.0 %) and T11-12 (41.6 %) had the highest prevalence rates. CONCLUSION: The prevalence of ossification of the ligamentum flavum was highest in the 50-59 years group, but also occurred in early years. OLF occurs more frequently in the lower than in the upper and middle thoracic regions and its prevalence increases with aging.

[Effect of tube voltage on digital chest radiograph for phantom and occupational exposed workers].

OBJECTIVE: To explore the image quality of different tube voltage on digital chest radiograph for contrast detail phantom (CDRAD2.0) and occupational exposed workers. METHODS: The CDRAD2.0 phantom DR images of high KV and different tube voltages were analyzed by 3 readers, image quality figure (IQF) were calculated and compared; at the same time 136 exposed workers were examined with high-kV and DR chest radiograph of different tube voltages. Contrast to high-KV images, 10 anatomic sites were scored .The image differences were compared between DR and high-kV. RESULTS: On CDRAD2.0 phantom, the IQF value of DR images in 3 readers were minimum in the condition of 120 kV, average value was 22.25. The analysis of variance in model with random effects, the mean IQF value of different tube voltage DR image had a significant difference (F = 13.775, P<0.01); By Dunnett t-tests analysis, the mean IQF value of DR image in 120 kV and high kilovoltage had no difference (t = -0.58, P = 0.979); On clinical cases, the DR image of 120 kV showed the closest anatomy to the high KV, the mean had no significant difference with 0 (P > 0.05) with single sample t test. CONCLUSION: On the CDRAD2.0 phantom or clinical exposed workers, the DR image quality of 120 kV tube voltage equals to high-KV basically.

[Effect of digital radiography processing parameters on digital chest radiograph for occupational exposed workers].

OBJECTIVE: To investigate the effect of different processing parameters of digital radiography (DR) on the image quality of digital chest radiograph in dust-exposed workers. METHODS: One hundred and five dust-exposed workers underwent both high-KV radiography and DR to obtain chest radiographs; the image processing parameters were set by the conventional processing method for digital chest radiograph (method A) and the processing method based on the special requirements of occupational diseases (method B). With the high-KV chest radiograph as the reference, the image qualities at 10 anatomic sites of DR image were graded. The images acquired by DR and high-KV radiography were compared, and the DR images acquired by methods A and B were also compared. RESULTS: For method A, the scores at the 10 anatomic sites of DR image were mostly 0 and +1, accounting for over 88%, and the mean score was 0.23 ∼ 0.65, there was a significant difference between the mean score of DR image and the score of high-KV image (P < 0.001). For method B, the scores at the 10 anatomic sites of DR image were mostly 0, accounting for over 65%, and the mean score was -0.01∼ +0.02 except at the pleura and chest wall; there was no significant difference between the mean score of DR image and the score of high-KV image (P > 0.05). There were significant differences in the scores at the 10 anatomic sites between the DR images acquired by methods A and B (P < 0.01). CONCLUSION: The DR images acquired based on different processing parameters are different. The quality of DR image acquired by the processing method based on the special requirements of occupational diseases is similar to that of high-KV image at the anatomic sites.

A brain structural connectivity biomarker for autism spectrum disorder diagnosis in early childhood.

Background: Autism spectrum disorder (ASD) is associated with altered brain development, but it is unclear which specific structural changes may serve as potential diagnostic markers, particularly in young children at the age when symptoms become fully established. Furthermore, such brain markers need to meet the requirements of precision medicine and be accurate in aiding diagnosis at an individual rather than only a group level. Objective: This study aimed to identify and model brain-wide differences in structural connectivity using diffusion tensor imaging (DTI) in young ASD and typically developing (TD) children. Methods: A discovery cohort including 93 ASD and 26 TD children and two independent validation cohorts including 12 ASD and 9 TD children from three different cities in China were included. Brain-wide (294 regions) structural connectivity was measured using DTI (fractional anisotropy, FA) together with symptom severity and cognitive development. A connection matrix was constructed for each child for comparisons between ASD and TD groups. Pattern classification was performed on the discovery dataset and the resulting model was tested on the two independent validation datasets. Results: Thirty-three structural connections showed increased FA in ASD compared to TD children and associated with both autistic symptom severity and impaired general cognitive development. The majority (29/33) involved the frontal lobe and comprised five different networks with functional relevance to default mode, motor control, social recognition, language and reward. Overall, classification achieved very high accuracy of 96.77% in the discovery dataset, and 91.67% and 88.89% in the two independent validation datasets. Conclusions: Identified structural connectivity differences primarily involving the frontal cortex can very accurately distinguish novel individual ASD from TD children and may therefore represent a robust early brain biomarker which can address the requirements of precision medicine.

Permanent implantation of iodine-125 seeds as a salvage therapy for recurrent head and neck carcinoma after radiotherapy.

Seventeen patients with head and neck recurrent carcinoma underwent (125)I seed implantation under CT or ultrasound guidance. The actuarial D90 of the (125)I seeds implanted was 90-160 Gy (median, 126 Gy). Median follow-up was 10 months (range, 3-48 months). The median local control time was 16 months; the 1- and 2-year local control rates were 66.5% and 49.9%, respectively. The 1- and 2-year survival rates were 51.3% and 38.5%, respectively (median, 16 months). None of the patients experienced grade 4 toxicity. (125)I seed implantation was a feasible and effective salvage treatment for patients with recurrent head and neck cancers.

Aggressive vertebral hemangioma of the thoracic spine without typical radiological appearance.

PURPOSE: Vertebral hemangioma (VH) is virtually vascular malformation, which is usually asymptomatic. Only 3.7 % of VH may become active and symptomatic, and 1 % may invade the spinal canal and/or paravertebral space. Treatment protocols for active or aggressive VHs are still in controversy. Reported treatments include radiotherapy, vertebroplasty, direct alcohol injection, embolization, surgery and a combination of these modalities. METHODS: A 41-year-old lady was presented with 18 month history of intermittent back pain. CT revealed T5 osteolytic lesion with epidural and paravertebral extension. The first CT guided biopsy yielded little information. RESULTS: Histopathological diagnosis of the second biopsy was VH. Vertebroplasty, posterior decompression and fixation were performed followed by postoperative radiotherapy. Her symptoms were resolved immediately after the operation. At 12 months follow-up, no recurrence was detected by CT with contrast enhancement. CONCLUSION: Surgical decompression, vertebroplasty and fixation are safe and effective for aggressive VH. More attention is needed in determining the algorithm for the diagnosis and treatment of aggressive VH.

[Hyperdense middle cerebral artery sign among patients with severe ischemic attack on ultra-early phase].

OBJECTIVE: To detect the frequency of hyperdense middle cerebral artery sign (HMCAS) among patients with severe ischemic attack (baseline NIHSS score ≥ 10) in the middle cerebral artery (MCA) territory within 3 hours after onset and compare the baseline characteristics and 90-day outcomes between the HMCAS (+) and HMCAS (-) groups. METHODS: A total of 43 patients were evaluated with baseline clinical characteristics, laboratory tests and brain CT/MRI. Follow-up evaluation at the end of 90 days included the modified Rankin Scale (mRS), the Bathel Index (BI) and death. RESULTS: Ten out of forty-three (23.3%) patients were HMCAS (+).The 24-hour NIHSSS of HMCAS (+) group was significantly higher than that of HMCAS (-) group (20.2 ± 5.4 vs. 14.8 ± 7.2, P= 0.037). The follow-up scanning confirmed that all HMCAS (+) patients had larger MCA territory infarction and the ratio of large MCA territory infarction was significantly higher than that of HMCAS (-) group (100% vs. 39.1%, P = 0.001). There were no significant differences in age, gender, risk factors, time intervals, laboratory tests, 90-day functional outcomes between the two groups. CONCLUSION: The HMCAS (+) patients had notable deterioration 24 hours after onset and had more severe MCA infarction. HMCAS detected by non-enhanced CT at admission should be considered as an alarming sign among patients with severe neurological deficit of MCA territory infarction.

BgNet: Classification of benign and malignant tumors with MRI multi-plane attention learning.

Objectives: To propose a deep learning-based classification framework, which can carry out patient-level benign and malignant tumors classification according to the patient's multi-plane images and clinical information. Methods: A total of 430 cases of spinal tumor, including axial and sagittal plane images by MRI, of which 297 cases for training (14072 images), and 133 cases for testing (6161 images) were included. Based on the bipartite graph and attention learning, this study proposed a multi-plane attention learning framework, BgNet, for benign and malignant tumor diagnosis. In a bipartite graph structure, the tumor area in each plane is used as the vertex of the graph, and the matching between different planes is used as the edge of the graph. The tumor areas from different plane images are spliced at the input layer. And based on the convolutional neural network ResNet and visual attention learning model Swin-Transformer, this study proposed a feature fusion model named ResNetST for combining both global and local information to extract the correlation features of multiple planes. The proposed BgNet consists of five modules including a multi-plane fusion module based on the bipartite graph, input layer fusion module, feature layer fusion module, decision layer fusion module, and output module. These modules are respectively used for multi-level fusion of patient multi-plane image data to realize the comprehensive diagnosis of benign and malignant tumors at the patient level. Results: The accuracy (ACC: 79.7%) of the proposed BgNet with multi-plane was higher than that with a single plane, and higher than or equal to the four doctors' ACC (D1: 70.7%, p=0.219; D2: 54.1%, p<0.005; D3: 79.7%, p=0.006; D4: 72.9%, p=0.178). Moreover, the diagnostic accuracy and speed of doctors can be further improved with the aid of BgNet, the ACC of D1, D2, D3, and D4 improved by 4.5%, 21.8%, 0.8%, and 3.8%, respectively. Conclusions: The proposed deep learning framework BgNet can classify benign and malignant tumors effectively, and can help doctors improve their diagnostic efficiency and accuracy. The code is available at https://github.com/research-med/BgNet.