Are current machine learning applications comparable to radiologist classification of degenerate and herniated discs and Modic change? A systematic review and meta-analysis.

INTRODUCTION: Low back pain is the leading contributor to disability burden globally. It is commonly due to degeneration of the lumbar intervertebral discs (LDD). Magnetic resonance imaging (MRI) is the current best tool to visualize and diagnose LDD, but places high time demands on clinical radiologists. Automated reading of spine MRIs could improve speed, accuracy, reliability and cost effectiveness in radiology departments. The aim of this review and meta-analysis was to determine if current machine learning algorithms perform well identifying disc degeneration, herniation, bulge and Modic change compared to radiologists. METHODS: A PRISMA systematic review protocol was developed and four electronic databases and reference lists were searched. Strict inclusion and exclusion criteria were defined. A PROBAST risk of bias and applicability analysis was performed. RESULTS: 1350 articles were extracted. Duplicates were removed and title and abstract searching identified original research articles that used machine learning (ML) algorithms to identify disc degeneration, herniation, bulge and Modic change from MRIs. 27 studies were included in the review; 25 and 14 studies were included multi-variate and bivariate meta-analysis, respectively. Studies used machine learning algorithms to assess LDD, disc herniation, bulge and Modic change. Models using deep learning, support vector machine, k-nearest neighbors, random forest and naïve Bayes algorithms were included. Meta-analyses found no differences in algorithm or classification performance. When algorithms were tested in replication or external validation studies, they did not perform as well as when assessed in developmental studies. Data augmentation improved algorithm performance when compared to models used with smaller datasets, there were no performance differences between augmented data and large datasets. DISCUSSION: This review highlights several shortcomings of current approaches, including few validation attempts or use of large sample sizes. To the best of the authors' knowledge, this is the first systematic review to explore this topic. We suggest the utilization of deep learning coupled with semi- or unsupervised learning approaches. Use of all information contained in MRI data will improve accuracy. Clear and complete reporting of study design, statistics and results will improve the reliability and quality of published literature.

Halo Femoral Traction for 1 Week Between Staged Anterior and Posterior Fusion Surgeries for Severe Adolescent Scoliosis Is Effective and Safe.

OBJECTIVE: To report the outcomes of halo femoral traction (HFT) used for 1 week between anterior release and definitive posterior fusion in adolescents with severe rigid scoliosis. METHODS: A retrospective single-center review of 22 consecutive patients (mean age at surgery, 14.1 years; range, 10.5-18.2 years; 17 girls) with severe, rigid scoliosis treated with anterior release, followed by HFT for 7 days prior to posterior instrumented fusion. Cobb angles were measured preoperatively, 1 week after anterior release and traction, after posterior fusion, and at a minimum 2-year follow-up. Complications were recorded. RESULTS: Mean preoperative Cobb angle was 97° (range, 80°-118°), correcting to 52° with anterior release and HFT and 31° after posterior fusion. This equated to a 68% deformity correction and was maintained at final follow-up. Three traction-related complications were experienced, including 1 case of neck pain and 2 cases of brachial plexopathy that resolved with traction weight reduction. CONCLUSIONS: Three-staged deformity correction using HFT for 1 week only offers gradual correction of the spine over sufficient time to optimize deformity correction yet minimizes neurologic dysfunction.

Nerve root injection cancellations because of incomplete anticoagulation management.

BACKGROUND: Computed tomography-guided steroid injection is a well-recognized, conservative treatment of localized spinal pain as a result of facet arthropathy and radiculopathy secondary to nerve root compression. An extremely rare complication is the development of an epidural haematoma with potential to cause permanent neurological damage, so anticoagulation at the time of procedure is contraindicated. Routinely injections are performed as an outpatient requiring the referring physician to implement a peri-procedural anticoagulation plan. Anecdotal experience suggested that cancellations were occurring as patients remained on anticoagulation at the time of their appointment. The authors therefore assessed the existing service against expected standards to identify the causes of cancellations and find ways to improve the service. AIMS: This audit aimed to identify the incidence of cancelled computed tomography-guided nerve root injections secondary to incorrect peri-procedural anticoagulation management, develop an intervention to help reduce the incidence of cancellations and then re-audit to assess the effect of the intervention. METHODS: The audit standard was that 100% of outpatients attending for computed tomography-guided nerve root and facet injections should have an appropriate anticoagulation plan implemented. Baseline data collection took place prospectively between 1 September and 30 November 2016. The study population was elective computed tomography-guided spinal nerve root and facet injections scheduled on the radiology information system at the authors' trust. Descriptive analysis was completed. The intervention involved a revised electronic request form being implemented with new compulsory fields concerning antiplatelets and anticoagulants. Re-audit post-intervention involved prospective data collection between 1 September and 30 November 2017 using the same methods. RESULTS: Baseline audit found that of three out of 55 (5%) patients had cancellations. On re-audit, there were 0 cancellations out of 93 patients. CONCLUSIONS: The new request form prevented 5% of patients referred for computed tomography-guided nerve root injection being cancelled because of incorrect anticoagulation management. Extrapolated over the year the potential savings through preventing lost activity are £3445.56.

Risk factors for and complications of distraction osteogenesis.

PURPOSE: Distraction osteogenesis is commonly used for limb deformities and reconstruction of bone defects with satisfactory outcome for the patients. However, it is associated with a risk of complications. The present study aims to assess the incidence of complications and to identify the risk factors that may predict distraction osteogenesis-related complications. MATERIALS AND METHODS: We retrospectively studied 63 patients (mean age 13.5 years; range 3-57 years) who had 74 distraction osteogenesis procedures from 2004 to 2009. A circular external fixator was used in 58 procedures, and a monolateral in 16 procedures. Fixator's time, days of treatment, lengthening percentage, bone healing index, distraction regenerate length and index, risk factors and complications were evaluated. The mean follow-up was 5 years (range 2-7 years). RESULTS: Complications occurred in 57 of the 74 procedures (77%); 70% were major complications and 30% were minor. Complications were more common in adults. Bone healing index, days of treatment and fixator's time were univariate predictors of complications. Bone healing index and adult age were the only multivariate predictors of complications. CONCLUSION: Adult age and bone healing index are the most important multivariate predictors of distraction osteogenesis-related complications. Routine follow-up after implant removal, selection of younger patients with minor risk factors and shorter fixator's time are necessary to reduce the rate of distraction osteogenesis-related complications.

Diving injuries of the cervical spine in amateur divers.

BACKGROUND CONTEXT: Diving injuries are the cause of potentially devastating trauma, primarily affecting the cervical spine. PURPOSE: Our purpose was to describe our experience with diving injuries treatment. STUDY DESIGN: Retrospective review. PATIENT SAMPLE: Twenty patients with diving injuries. OUTCOME MEASURES: Using the American Spinal Injury Association (ASIA) impairment scales as the primary outcome measure, the patients' neurological status before and after treatment was assessed. In this way we were able to draw conclusions about neurological improvement or deterioration in response to conservative or operative treatment. METHODS: We retrospectively reviewed 20 patients with diving injuries of the cervical spine who were admitted to our institute over a 34-year period from 1970 until 2004. RESULTS: The typical patient profile was of a young, healthy, athletic male who suffered an injury to the cervical spine after diving into shallow water. The number of cases corresponds to 2.6% of all admitted cervical spine injuries. All injures occurred between May and September. The most commonly fractured vertebrae were C5 and C6. Four patients were treated operatively and 16 conservatively. The indications for surgical treatment were posttraumatic instability and persistent neurological deficit. The mean follow-up of the patients was 17 years. Five patients died within the first month of their hospitalization and 1 patient died 1 year after his injury. Of the 14 patients who were available for follow-up 5 years past injury time, 6 improved neurologically and 8 remained unchanged in relation to their neurology upon admission. Of the 11 patients who were available for follow-up 10 years past injury time, 9 remained neurologically unchanged, 1 deteriorated, and 1 improved in relation to their neurology in the 5-year follow-up. CONCLUSION: Diving injuries of the cervical spine demonstrate high mortality and morbidity rates. Recovery depends on the severity of the initial neurological damage. Conservative treatment is justified in specific patients and can lead to improvement of the initial neurological deficit.