Neoadjuvant chemotherapy (NAC) is the primary method to reduce the burden of tumor and metastasis; in the treatment of breast cancer, it may provide additional opportunities for breast-conserving surgery. Preoperative assessment of pathological complete response (PCR) to NAC is important for developing individualized treatment approaches and predicting patient prognosis. Compared to magnetic resonance imaging (MRI) and mammography, ultrasonography (US) has the advantages of simplicity, flexibility, and real-time imaging. Moreover, it does not require radiation and can provide multi-time acquisition of the tumor during NAC treatment. Recently, deep learning radiomics models based on multi-time-point US images for the prediction of NAC effectiveness have been proposed. To further improve the prediction performance, we carefully designed four supporting modules for our proposed dual-input transformer (DiT): isolated tokens-to-token patch embedding module, shared position embedding, time embedding, and weighted average pooling feature representation modules. The design of each module considers the characteristics of the US images at multiple time points. We validated our model on our retrospective US dataset composed of 484 cases from two centers whose consistency is not sufficiently high. Patients were allocated to training (n = 297), validation (n = 99), and external test (n = 88) sets. The results show that our model can achieve better performance than the Siamese CNN and the standard tokens-to-token vision transformer without using multi-time-point images. The ablation study also proved the effectiveness of each module designed for DiT.
13 January 2016 in Journal of Orthopaedic Surgery and Research
The purpose of this study was to investigate the clinical outcomes of anterior cervical discectomy and fusion (ACDF) or anterior cervical corpectomy and fu
16 April 2018 in Journal of Orthopaedic Surgery and Research
The current study aimed to discuss the prevalence and surgical outcomes of the asymmetrical Modic changes and identify its correlative factors by multivariate logistic regression analysis. Two hundred seventy-eight patients with single-level Modic changes and nerve compression symptoms were reviewed retrospectively from January 2008 to January 2015. 1.5-T MRI was performed to determine the Modic changes. Multivariate logistic regression analysis was used to identify the correlative factors of asymmetrical Modic changes. Surgeries were performed according to the surgical indications. The outcomes were recorded by Japanese Orthopaedic Association (JOA) score, Neck Disability Index (NDI) score, and recovery rate. Asymmetrical Modic changes were observed in 76 patients (27.34%) with 4 type 1, 69 type 2, and 3 type 3. C5/6 was the most frequently affected segment with 39 patients showing signal changes on MRI. Statistically significant difference was showed in conservative rehabilitation rate between two groups (p = 0.043). Multiple logistic regression analysis identified disc herniation and neurological symptoms as correlative factors of asymmetrical Modic changes, and the adjusted odds ratios (95% CI) were 2.079 (1.348–3.208) and 0.231 (0.143–0.373) respectively. No statistically significant difference was found in JOA scores and NDI scores between the two kinds of Modic changes. C5/6 was the most commonly affected level by Modic changes. Disc herniation and nerve root compression symptom were more closely correlated with asymmetrical Modic changes than conventional Modic changes. Asymmetrical Modic changes indicated poor result in conservative treatment; however, the final operation rate was similar between the two kinds of Modic changes. The outcomes of surgical treatment were satisfactory both in patients with asymmetrical Modic changes and conventional Modic changes.
20 September 2016 in Journal of Orthopaedic Surgery and Research
The purpose of this study was to investigate whether uncovertebral joint ossification was a risk factor for axial symptoms (AS) after cervical disc arthroplasty (CDA). This retrospective study included 52 consecutive patients who underwent CDA for single-level cervical disc disease. To examine possible risk factors for AS after CDA, univariate and multivariate logistic regression analyses were conducted to compare data from the patients with and without AS (the AS and no-AS groups, respectively). Among the 52 patients examined, AS were observed in 24 patients (46.2 %), including a stiff neck (n = 11), neck pain and dullness (n = 10), and shoulder pain (n = 3). Uncovertebral joint ossification was detected in 22 (42.3 %) patients, including 17 patients in the AS group and 5 patients in the no-AS group. Clinical outcome improved during the follow-up period for the AS group. According to multivariate logistic regression analysis, uncovertebral joint ossification, cervical kyphosis, and range of motion (ROM) at the index level were identified as significant risk factors for AS after CDA. Satisfactory clinical outcomes were observed following CDA for the treatment of single-level cervical disc disease in the present cohort. In addition, uncovertebral joint ossification, cervical kyphosis, and ROM at the index level were found to affect the incidence of AS after CDA.