Detection and localisation of hip fractures on anteroposterior radiographs with artificial intelligence: proof of concept.

AIM: To investigate the feasibility of applying a deep convolutional neural network (CNN) for detection/localisation of acute proximal femoral fractures (APFFs) on hip radiographs. MATERIALS AND METHODS: This study had institutional review board approval. Radiographs of 307 patients with APFFs and 310 normal patients were identified. A split ratio of 3/1/1 was used to create training, validation, and test datasets. To test the validity of the proposed model, a 20-fold cross-validation was performed. The anonymised images from the test cohort were shown to two groups of radiologists: musculoskeletal radiologists and diagnostic radiology residents. Each reader was asked to assess if there was a fracture and localise it if one was detected. The area under the receiver operator characteristics curve (AUC), sensitivity, and specificity were calculated for the CNN and readers. RESULTS: The mean AUC was 0.9944 with a standard deviation of 0.0036. Mean sensitivity and specificity for fracture detection was 97.1% (81.5/84) and 96.7% (118/122), respectively. There was good concordance with saliency maps for lesion identification, but sensitivity was lower for characterising location (subcapital/transcervical, 84.1%; basicervical/intertrochanteric, 77%; subtrochanteric, 20%). Musculoskeletal radiologists showed a sensitivity and specificity for fracture detection of 100% and 100% respectively, while residents showed 100% and 96.8%, respectively. For fracture localisation, the performance decreased slightly for human readers. CONCLUSION: The proposed CNN algorithm showed high accuracy for detection of APFFs, but the performance was lower for fracture localisation. Overall performance of the CNN was lower than that of radiologists, especially in localizing fracture location.

A prospective randomized clinical trial to evaluate the impact of intraoperative ventilation with high oxygen content on the extent of postoperative pneumocephalus in patients undergoing craniotomies.

Purpose of the article: To determine whether intraoperative ventilation with pure oxygen during the last stage of surgery reduces the occurrence and volume of postoperative pneumocephalus when compared to conventional air/oxygen mixture in patients undergoing craniotomy. MATERIAL AND METHODS: prospective randomized single-blinded study to compare the rate of occurrence and volume of postoperative pneumocephalus in patients undergoing craniotomy receiving intraoperative ventilation with pure oxygen (Group B) versus a conventional air/oxygen 1:1 mixture (Group A) during the last stage of surgery. This trial was registered in ClinicalTrials.gov #NCT02722928, protocol number 2015H0032. RESULTS: One hundred patients were randomized into group 'A' and group 'B'. Seventy patients were included in the final analysis with 39 patients allocated in group 'A' and 31 patients in group 'B'. Median and IQR were used for postoperative penumocephalus volume. Group A: 9.65 [3.61-23.20]; Group B: 7.06 [2.70-20.1]. Our study showed no prophylactic effect on postoperative pneumocephalus volume when using mechanical ventilation with higher oxygen concentrations than the standard FiO2 during the last stage of surgery in patients undergoing craniotomy (p = .47). No statistical difference was found in SICU LOS between groups (median 1,380 min [group A] versus 1,524 min [group B]; p = .18). CONCLUSION: The use of intraoperative mechanical ventilation with pure oxygen was not associated with a prophylactic effect on the occurrence and extent of postoperative pneumocephalus in our patient setting. Published literature describing the extent of postoperative pneumocephalus is limited or highly variable among institutions.

An Artificial Intelligence Tool for Clinical Decision Support and Protocol Selection for Brain MRI.

BACKGROUND AND PURPOSE: Protocolling, the process of determining the most appropriate acquisition parameters for an imaging study, is time-consuming and produces variable results depending on the performing physician. The purpose of this study was to assess the potential of an artificial intelligence-based semiautomated tool in reducing the workload and decreasing unwarranted variation in the protocolling process. MATERIALS AND METHODS: We collected 19,721 MR imaging brain examinations at a large academic medical center. Criterion standard labels were created using physician consensus. A model based on the Long Short-Term Memory network was trained to predict the most appropriate protocol for any imaging request. The model was modified into a clinical decision support tool in which high-confidence predictions, determined by the values the model assigns to each possible choice, produced the best protocol automatically and low confidence predictions provided a shortened list of protocol choices for review. RESULTS: The model achieved 90.5% accuracy in predicting the criterion standard labels and demonstrated higher agreement than the original protocol assignments, which achieved 85.9% accuracy (κ = 0.84 versus 0.72, P value < .001). As a clinical decision support tool, the model automatically assigned 70% of protocols with 97.3% accuracy and, for the remaining 30% of examinations, achieved 94.7% accuracy when providing the top 2 protocols. CONCLUSIONS: Our model achieved high accuracy on a standard based on physician consensus. It showed promise as a clinical decision support tool to reduce the workload by automating the protocolling of a sizeable portion of examinations while maintaining high accuracy for the remaining examinations.

[The neuropathology of neonatal period: analysis of 1616 autopsies]. / A neuropatologia no período neonatal: análise de 1616 casos de necropsia.

The lesions of the central nervous system represent an important cause of morbid-mortality in the neonatal period. This is due to the vulnerability of the brain to several adverse conditions during gestation and after birth. This study analyses the prevalence and pattern of central nervous system lesions in neonates autopsied at Hospital de Clínicas - Curitiba. There were 5743 pediatric autopsies performed in the Sector of Anatomic Pathology from 1960 to 1995 with 2049 cases corresponding to death during neonatal period. These later autopsies were reviewed and all cases with central nervous system lesions were selected and classified according to sex, age and pattern of central nervous system lesion. The central nervous system was affected in 1616 (78,87%) of neonatal autopsies and there was predominance of intracerebral hemorrhages (73,39%), congenital malformations (4,27%) and infections (3,59%). The hypoxic hemorrhages are the most prevalent central nervous system lesions in the neonatal period, affecting mainly premature babies. There was predominance of central nervous system malformations in the female neonates.