Thoracic spinous process nonunion as an unusual cause of back pain: a case report and review of the literature.

BACKGROUND: Purely isolated spinous processes fractures are rare and are usually treated conservatively, although a few authors have reported cases of nonunion that ultimately required surgical resection. CASE PRESENTATION: We present a case of an isolated T6 spinous process pseudoarthrosis that was treated by surgical resection of the tip of the spinous process. A 34-year-old Caucasian male patient was complaining of mid-thoracic back pain without neurologic impairment more than 2 years after an isolated spinous process fracture. Magnetic Resonance Imaging (MRI) and Single Photon Emission Computed Tomography (SPECT) revealed a nonunion. We performed a resection without further complication. CONCLUSION: Although spinous process nonunions may in some cases be well tolerated, surgical resection appears to be a reliable option in case of persistent symptoms. This illustrated case shows the description of an isolated thoracic spinous process nonunion and its surgical treatment.

Interspinalis-longissimus mid-line approach for thoracic and thoracolumbar minimally invasive fixation: Technical note.

The authors present a new minimally invasive posterior thoracic and upper lumbar spinal approach, using midline-staged incision with bilateral dissection via the natural muscular cleavage plane between the spinalis and longissimus muscles for pedicle fixation. The aim was to clarify anatomy and compare results between this technique and the conventional paraspinal approach. An anatomic MRI study determined the position of the superficial aponeurosis of the spinalis-longissimus intermuscular cleavage plane with respect to the midline. The surgical procedure is described. A radiological and clinical study evaluated 24 patients divided into two comparative groups, operated on via the medial intermuscular or the paraspinal approach, for isolated A3 and B2 Magerl fracture between T2 and L2, without neurologic involvement. Pain, postoperative complications, reduction quality and esthetic satisfaction were assessed. MRI located the cleavage plane, which was consistently 11.01±0.9mm (range, 9.4-12.3mm) from the midline. In the "medial approach" group, hospital stay was significantly shorter, by 3 days, at 4±1.9 (range, 2-8 days) (p: 0.005); pain on VAS was 2 points lower at postoperative day 1 (2.1±1.4; range,≤1-4; p:0.032), and lower at discharge (0.3±0.6; range, 0-1; p: 0.001). The scar was considered esthetically very satisfactory. This easily performed and reproducible medial intermuscular approach showed low iatrogenesis, significantly improving postoperative course.

Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autism.

Background: Autism spectrum disorder (ASD) diagnosis can be delayed due in part to the time required for administration of standard exams, such as the Autism Diagnostic Observation Schedule (ADOS). Shorter and potentially mobilized approaches would help to alleviate bottlenecks in the healthcare system. Previous work using machine learning suggested that a subset of the behaviors measured by ADOS can achieve clinically acceptable levels of accuracy. Here we expand on this initial work to build sparse models that have higher potential to generalize to the clinical population. Methods: We assembled a collection of score sheets for two ADOS modules, one for children with phrased speech (Module 2; 1319 ASD cases, 70 controls) and the other for children with verbal fluency (Module 3; 2870 ASD cases, 273 controls). We used sparsity/parsimony enforcing regularization techniques in a nested cross validation grid search to select features for 17 unique supervised learning models, encoding missing values as additional indicator features. We augmented our feature sets with gender and age to train minimal and interpretable classifiers capable of robust detection of ASD from non-ASD. Results: By applying 17 unique supervised learning methods across 5 classification families tuned for sparse use of features and to be within 1 standard error of the optimal model, we find reduced sets of 10 and 5 features used in a majority of models. We tested the performance of the most interpretable of these sparse models, including Logistic Regression with L2 regularization or Linear SVM with L1 regularization. We obtained an area under the ROC curve of 0.95 for ADOS Module 3 and 0.93 for ADOS Module 2 with less than or equal to 10 features. Conclusions: The resulting models provide improved stability over previous machine learning efforts to minimize the time complexity of autism detection due to regularization and a small parameter space. These robustness techniques yield classifiers that are sparse, interpretable and that have potential to generalize to alternative modes of autism screening, diagnosis and monitoring, possibly including analysis of short home videos.