External validation of a convolutional neural network artificial intelligence tool to predict malignancy in pulmonary nodules.

BACKGROUND: Estimation of the risk of malignancy in pulmonary nodules detected by CT is central in clinical management. The use of artificial intelligence (AI) offers an opportunity to improve risk prediction. Here we compare the performance of an AI algorithm, the lung cancer prediction convolutional neural network (LCP-CNN), with that of the Brock University model, recommended in UK guidelines. METHODS: A dataset of incidentally detected pulmonary nodules measuring 5-15 mm was collected retrospectively from three UK hospitals for use in a validation study. Ground truth diagnosis for each nodule was based on histology (required for any cancer), resolution, stability or (for pulmonary lymph nodes only) expert opinion. There were 1397 nodules in 1187 patients, of which 234 nodules in 229 (19.3%) patients were cancer. Model discrimination and performance statistics at predefined score thresholds were compared between the Brock model and the LCP-CNN. RESULTS: The area under the curve for LCP-CNN was 89.6% (95% CI 87.6 to 91.5), compared with 86.8% (95% CI 84.3 to 89.1) for the Brock model (p≤0.005). Using the LCP-CNN, we found that 24.5% of nodules scored below the lowest cancer nodule score, compared with 10.9% using the Brock score. Using the predefined thresholds, we found that the LCP-CNN gave one false negative (0.4% of cancers), whereas the Brock model gave six (2.5%), while specificity statistics were similar between the two models. CONCLUSION: The LCP-CNN score has better discrimination and allows a larger proportion of benign nodules to be identified without missing cancers than the Brock model. This has the potential to substantially reduce the proportion of surveillance CT scans required and thus save significant resources.

Predictors of spontaneous intracerebral hemorrhage mortality: a community-based study in Brno, Czech Republic.

BACKGROUND AND OBJECTIVE: Intracerebral hemorrhage (ICH) is a serious medical condition with high mortality. However, factors leading to long-term mortality after ICH are largely unclear. The aim of this community-based study is to assess predictors of long-term mortality after spontaneous ICH. METHODS: We identified all patients admitted with spontaneous ICH to hospitals with a certified stroke unit in Brno, the second largest city in the Czech Republic (CR), in 2011, the year of the Czech Population and Housing Census. We reviewed their medical records for risk factors, radiographic parameters, and measures of post-stroke neurological deficit [National Institutes of Health Stroke Scale (NIHSS)]. Using the dates of death from the Czech National Mortality Register, we calculated mortality at 30 days, six months, one year, and three years after the ICH. Multivariate analysis with forward stepwise logistic regression was performed to determine independent predictors of mortality (p < 0.05). RESULTS: In 2011, 1086 patients with stroke were admitted to the four stroke-certified hospitals in Brno, CR. Of these, 134 had spontaneous ICH, with complete data available in 93 of them entering the final analysis. The mortality at 30 days, 6 months, 1 year, and 3 years post-ICH was 34%, 47%, 51%, and 63%, respectively. The mortality was highest in the first few days post-event, with 50% of patients dying in 255 days and average survival being 884 ± 90 days. Both NIHSS and modified ICH (MICH) score showed to be strong and reliable predictors of short- as well as long-term mortality; the risk of death post-ICH increased with older age and size of ICH. Other risk factors contributing to higher, primarily shorter-term mortality included history of cardiac failure, myocardial infarction, or atrial fibrillation. CONCLUSIONS: In our community-based study, we found that severity of neurological deficit at admission (NIHSS), combined with age and size of ICH, well predicted short- as well as long-term mortality after spontaneous ICH. A history of cardiac failure, myocardial infarction, or atrial fibrillation at presentation were also predictors of mortality, underscoring the need for optimal cardiac management in patients with ICH.

Simple Microsurgical Extirpation as a Method of Choice in Treating Symptomatic Spinal Facet Joint Cysts: A Retrospective Case Series.

BACKGROUND: Facet joint cysts (FJCs) are a cause of radiculopathy, back pain, cauda equina syndrome, and/or claudication. They mostly affect the lumbar spine in the elderly, prevailing in women, and are connected to spinal degeneration and instability. We aimed to evaluate the safety and efficacy of open surgical decompression and cyst excision without a subsequent fusion. METHODS: We evaluated neurologic symptoms and potential signs of spinal instability on preoperative versus postoperative radiological examination. We performed a histological assessment of the extirpated cysts. Statistical analysis was then performed. RESULTS: Of a total of 66 patients, 44 were included in the present study. The average age was 61.2 years. Female patients prevailed (61.4%). The mean follow-up was 5.3 years. The segment most frequently affected by a FJC was L4-L5 (65.9%). Cyst resection led to significant relief from neurologic symptoms for most patients. Thus, 95.5% of our patients reported their postoperative outcome as excellent. Preoperatively, 43.2% and 47.4% of patients had had radiographic signs of instability on magnetic resonance imaging and signs of spondylolisthesis on dynamic radiographs in the operated segment, respectively, with 54.5% manifesting signs of spondylolisthesis in the same segment on a postoperative dynamic radiograph. Despite spondylolisthesis progression, no patient required reoperation. Histologically, pseudocysts without synovium were more frequent than were synovial cysts. CONCLUSIONS: Simple FJC extirpation is a safe and effective method for resolving radicular symptoms, with excellent long-term outcomes. It does not lead to the development of clinically significant spondylolisthesis in the operated segment; thus, no supplementary fusion with instrumented stabilization is required.

Long-Term Outcome After Midline Lumbar Fusion for the Treatment of Lumbar Spine Instability Due to Degenerative Disease.

BACKGROUND: Midline lumbar fusion (MIDLF) is one promising new surgical technique that has been developed to minimize perioperative damage to the paravertebral stabilizing musculotendinous system. The aim of this study was to assess long-term clinical and radiological effects of MIDLF. METHODS: This prospective cohort study evaluated patients who underwent MIDLF for degenerative spinal instability. Clinical and radiological examinations were performed before and after surgery. Perioperative and postoperative complications were recorded. Follow-up was 2 years. P ≤ 0.05 was considered statistically significant. RESULTS: The study included 64 patients (mean age 58.9 ± 10.7 years; 41 women [64.1%]). The most frequent indication for MIDLF was degenerative spondylolisthesis grade I (28 cases [43.8%]); the prevalent spinal segment to be fused was L4-L5 (35 cases [54.7%]). Mean duration of surgery was 148.2 ± 28.9 minutes. Relief of low back pain and leg pain was significant and stable in the postoperative period as assessed by visual analog scale (P < 0.001). Of patients, 86.9% reported fair, good, or excellent outcomes in terms of pain relief based on MacNab score 2 years after surgery. Patients' level of function in activities of daily living improved significantly based on Oswestry Disability Index score: from 66.8 ± 9.8 before surgery to 33.9 ± 16.5 2 years after surgery (P < 0.001). X-rays and computed tomography at 12 months showed interbody fusion in 46 cases (73.4%), inconclusive results in 13 cases (20.3%), and no fusion in 4 cases (6.3%). No damage to neural or vascular structures and no failure of hardware or screw loosening were recorded. CONCLUSIONS: MIDLF is a safe, efficient method for surgical treatment of lumbar spine instability. Its limited invasiveness contributes to better preservation of paravertebral muscles and enhanced postoperative spinal stability.

Mirror Therapy in Stroke Rehabilitation: Current Perspectives.

In contrast to varied therapy approaches, mirror therapy (MT) can be used even in completely plegic stroke survivors, as it uses visual stimuli for producing a desired response in the affected limb. MT has been studied to have effects not just on motor impairments but also on sensations, visuospatial neglect, and pain after stroke. This paper attempts to systematically review and present the current perspectives on mirror therapy and its application in stroke rehabilitation, and dosage, feasibility and acceptability in stroke rehabilitation. An electronic database search across Google, PubMed, Web of Science, etc., generated 3871 results. After screening them based on the inclusion and exclusion criteria, we included 28 studies in this review. The data collected were divided on the basis of application in stroke rehabilitation, modes of intervention delivery, and types of control and outcome assessment. We found that most studies intervened for upper limb motor impairments post stroke. Studies were equally distributed between intervention in chronic and acute phases post stroke with therapy durations lasting between 1 and 8 weeks. MT showed definitive motor and sensory improvements although the extent of improvements in sensory impairments and hemineglect is limited. MT proves to be an effective and feasible approach to rehabilitate post-stroke survivors in the acute, sub-acute, and chronic phases of stroke, although its long-term effects and impact on activities of daily living need to be analysed extensively.