Impact of Structured Reporting of Skeletal Survey in Skeletal Dysplasia: A Single Institution Experience.

Background Structured reporting has the advantages of reducing ambiguity in written radiology reports with greater uniformity and comparability of reports amongst different institutes. It has multiple facets: structured format, structured content, and standardized language. While structured reporting initiative has been used in various radiology subspecialties such as oncology, cardiothoracic, abdominal and interventional radiology; skeletal dysplasia is a domain that remains largely untouched by this concept. Purpose To evaluate the impact of structured reporting in skeletal dysplasia. Methods and Materials This was an ethically approved pragmatic clinical trial. A defined number (75) of clinically diagnosed and/or genetically confirmed skeletal dysplasia radiographs were evaluated by two radiologists (reader A and reader B) with 5-and 7-years' experience in general radiology, respectively. A pre-defined structured reporting format for skeletal dysplasia was used as an interventional tool. Both the readers interpreted the radiographs before and after the training session. In addition to diagnosis, diagnostic confidence was noted using a semiquantitative scale. Improvement in diagnostic accuracy and diagnostic confidence after training were assessed. McNemar's test was used to assess the statistical significance of difference in proportion of correct diagnoses in pre- and post-education phases. An interrater reliability analysis using the Kappa statistic was performed to determine interobserver agreement between readers both in pre- and post-education phases. Results In the post-education phase, the proportion of accurate diagnosis improved from 48% (36/75) to 64% (48/75) for reader A, and from 44% (33/75) to 60% (45/75) for reader B as compared with the pre-education phase. Amongst the cases with a correct radiologic diagnosis, an increase in diagnostic confidence was noted in 18 cases for reader A, and 15 cases for reader B. In none of the cases, there was a reduction in diagnostic confidence after training. A McNemar's test determined that there was a statistically significant difference in the proportion of correct diagnoses in pre- and post-education phases, p < 0.001. The interobserver agreement between the readers was found to increase from Kappa = 0.33 ( p = 0.004) using non-structured reporting in pre-education phase to Kappa = 0.46 ( p < 0.001) using structured reporting in the post-education phase. Conclusion A structured reporting of skeletal survey can improve accuracy and confidence in diagnosing skeletal dysplasia.

An open-label non-inferiority randomised control trial comparing nebulised amphotericin B with oral itraconazole in patients with pulmonary aspergilloma.

BACKGROUND: Pulmonary aspergilloma (PA) is a common complication seen in patients with pulmonary tuberculosis sequelae. Antifungal therapy, including oral azoles, is commonly used though only surgical resection offers curative benefit. Local administration of amphotericin B, like intracavitary instillation, has been effective in aspergilloma patients though nebulised amphotericin B (nAB) has never been formally assessed. OBJECTIVE: The aim of this prospective, non-inferior, open-label, randomised control trial is to evaluate the efficacy and safety of nebulised amphotericin B compared to oral itraconazole therapy in the treatment of PA. PATIENTS/METHODS: Diagnosed cases of PA (n=33) were randomised into the control group receiving oral itraconazole (n=18) and intervention group receiving nebulised amphotericin B (n = 15). Response to treatment was assessed both clinically and radiologically at the end 6 months. RESULTS AND CONCLUSION: The number of patients showing overall improvement at the end of 6 months in the control arm(oral itraconazole) vs intervention arm(nebulised amphotericin B) was 65% (95% CI 38.3-85.8) and 67%(95% CI 38.4%-88.2%), respectively, in the intention-to-treat and 79% (95% CI 49.2%-95.3%), and 65% (95% CI 38.4%-88.2%), respectively, in the per-protocol analysis. While there was no statistically significant difference between the intervention and control arm in both the analyses, non-inferiority was shown in the per-protocol but not in the intention-to-treat analysis. No major adverse events were noted in either group; however, a significant proportion of patients receiving nAB reported minor cough (40%), which, however, did not lead to discontinuation of therapy in any patients. Nebulised amphotericin B can be an effective therapeutic option for pulmonary aspergilloma patients.

Artificial Intelligence-assisted chest X-ray assessment scheme for COVID-19.

OBJECTIVES: To study whether a trained convolutional neural network (CNN) can be of assistance to radiologists in differentiating Coronavirus disease (COVID)-positive from COVID-negative patients using chest X-ray (CXR) through an ambispective clinical study. To identify subgroups of patients where artificial intelligence (AI) can be of particular value and analyse what imaging features may have contributed to the performance of AI by means of visualisation techniques. METHODS: CXR of 487 patients were classified into [4] categories-normal, classical COVID, indeterminate, and non-COVID by consensus opinion of 2 radiologists. CXR which were classified as "normal" and "indeterminate" were then subjected to analysis by AI, and final categorisation provided as guided by prediction of the network. Precision and recall of the radiologist alone and radiologist assisted by AI were calculated in comparison to reverse transcriptase-polymerase chain reaction (RT-PCR) as the gold standard. Attention maps of the CNN were analysed to understand regions in the CXR important to the AI algorithm in making a prediction. RESULTS: The precision of radiologists improved from 65.9 to 81.9% and recall improved from 17.5 to 71.75 when assistance with AI was provided. AI showed 92% accuracy in classifying "normal" CXR into COVID or non-COVID. Analysis of attention maps revealed attention on the cardiac shadow in these "normal" radiographs. CONCLUSION: This study shows how deployment of an AI algorithm can complement a human expert in the determination of COVID status. Analysis of the detected features suggests possible subtle cardiac changes, laying ground for further investigative studies into possible cardiac changes. KEY POINTS: • Through an ambispective clinical study, we show how assistance with an AI algorithm can improve recall (sensitivity) and precision (positive predictive value) of radiologists in assessing CXR for possible COVID in comparison to RT-PCR. • We show that AI achieves the best results in images classified as "normal" by radiologists. We conjecture that possible subtle cardiac in the CXR, imperceptible to the human eye, may have contributed to this prediction. • The reported results may pave the way for a human computer collaboration whereby the expert with some help from the AI algorithm achieves higher accuracy in predicting COVID status on CXR than previously thought possible when considering either alone.

Spine radiograph in dysplasias: A pictorial essay.

Spine radiograph is an essential component of a skeletal survey. It provides important diagnostic clues to various types of skeletal dysplasia. In some conditions, a spine radiograph alone may be diagnostic and characteristic; but mostly, it yields more value as a part of the complete skeletal survey. In this article we will discuss about a few common lethal and non-lethal skeletal dysplasias and their characteristic imaging findings; primarily focusing on the spine radiograph.