Using Artificial Intelligence to Diagnose Osteoporotic Vertebral Fractures on Plain Radiographs.

Osteoporotic vertebral fracture (OVF) is a risk factor for morbidity and mortality in elderly population, and accurate diagnosis is important for improving treatment outcomes. OVF diagnosis suffers from high misdiagnosis and underdiagnosis rates, as well as high workload. Deep learning methods applied to plain radiographs, a simple, fast, and inexpensive examination, might solve this problem. We developed and validated a deep-learning-based vertebral fracture diagnostic system using area loss ratio, which assisted a multitasking network to perform skeletal position detection and segmentation and identify and grade vertebral fractures. As the training set and internal validation set, we used 11,397 plain radiographs from six community centers in Shanghai. For the external validation set, 1276 participants were recruited from the outpatient clinic of the Shanghai Sixth People's Hospital (1276 plain radiographs). Radiologists performed all X-ray images and used the Genant semiquantitative tool for fracture diagnosis and grading as the ground truth data. Accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were used to evaluate diagnostic performance. The AI_OVF_SH system demonstrated high accuracy and computational speed in skeletal position detection and segmentation. In the internal validation set, the accuracy, sensitivity, and specificity with the AI_OVF_SH model were 97.41%, 84.08%, and 97.25%, respectively, for all fractures. The sensitivity and specificity for moderate fractures were 88.55% and 99.74%, respectively, and for severe fractures, they were 92.30% and 99.92%. In the external validation set, the accuracy, sensitivity, and specificity for all fractures were 96.85%, 83.35%, and 94.70%, respectively. For moderate fractures, the sensitivity and specificity were 85.61% and 99.85%, respectively, and 93.46% and 99.92% for severe fractures. Therefore, the AI_OVF_SH system is an efficient tool to assist radiologists and clinicians to improve the diagnosing of vertebral fractures. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Comparison of Chest Radiograph Captions Based on Natural Language Processing vs Completed by Radiologists.

Importance: Artificial intelligence (AI) can interpret abnormal signs in chest radiography (CXR) and generate captions, but a prospective study is needed to examine its practical value. Objective: To prospectively compare natural language processing (NLP)-generated CXR captions and the diagnostic findings of radiologists. Design, Setting, and Participants: A multicenter diagnostic study was conducted. The training data set included CXR images and reports retrospectively collected from February 1, 2014, to February 28, 2018. The retrospective test data set included consecutive images and reports from April 1 to July 31, 2019. The prospective test data set included consecutive images and reports from May 1 to September 30, 2021. Exposures: A bidirectional encoder representation from a transformers model was used to extract language entities and relationships from unstructured CXR reports to establish 23 labels of abnormal signs to train convolutional neural networks. The participants in the prospective test group were randomly assigned to 1 of 3 different caption generation models: a normal template, NLP-generated captions, and rule-based captions based on convolutional neural networks. For each case, a resident drafted the report based on the randomly assigned captions and an experienced radiologist finalized the report blinded to the original captions. A total of 21 residents and 19 radiologists were involved. Main Outcomes and Measures: Time to write reports based on different caption generation models. Results: The training data set consisted of 74 082 cases (39 254 [53.0%] women; mean [SD] age, 50.0 [17.1] years). In the retrospective (n = 8126; 4345 [53.5%] women; mean [SD] age, 47.9 [15.9] years) and prospective (n = 5091; 2416 [47.5%] women; mean [SD] age, 45.1 [15.6] years) test data sets, the mean (SD) area under the curve of abnormal signs was 0.87 (0.11) in the retrospective data set and 0.84 (0.09) in the prospective data set. The residents' mean (SD) reporting time using the NLP-generated model was 283 (37) seconds-significantly shorter than the normal template (347 [58] seconds; P < .001) and the rule-based model (296 [46] seconds; P < .001). The NLP-generated captions showed the highest similarity to the final reports with a mean (SD) bilingual evaluation understudy score of 0.69 (0.24)-significantly higher than the normal template (0.37 [0.09]; P < .001) and the rule-based model (0.57 [0.19]; P < .001). Conclusions and Relevance: In this diagnostic study of NLP-generated CXR captions, prior information provided by NLP was associated with greater efficiency in the reporting process, while maintaining good consistency with the findings of radiologists.

Development and multicenter validation of chest X-ray radiography interpretations based on natural language processing.

Background: Artificial intelligence can assist in interpreting chest X-ray radiography (CXR) data, but large datasets require efficient image annotation. The purpose of this study is to extract CXR labels from diagnostic reports based on natural language processing, train convolutional neural networks (CNNs), and evaluate the classification performance of CNN using CXR data from multiple centers. Methods: We collected the CXR images and corresponding radiology reports of 74,082 subjects as the training dataset. The linguistic entities and relationships from unstructured radiology reports were extracted by the bidirectional encoder representations from transformers (BERT) model, and a knowledge graph was constructed to represent the association between image labels of abnormal signs and the report text of CXR. Then, a 25-label classification system were built to train and test the CNN models with weakly supervised labeling. Results: In three external test cohorts of 5,996 symptomatic patients, 2,130 screening examinees, and 1,804 community clinic patients, the mean AUC of identifying 25 abnormal signs by CNN reaches 0.866 ± 0.110, 0.891 ± 0.147, and 0.796 ± 0.157, respectively. In symptomatic patients, CNN shows no significant difference with local radiologists in identifying 21 signs (p > 0.05), but is poorer for 4 signs (p < 0.05). In screening examinees, CNN shows no significant difference for 17 signs (p > 0.05), but is poorer at classifying nodules (p = 0.013). In community clinic patients, CNN shows no significant difference for 12 signs (p > 0.05), but performs better for 6 signs (p < 0.001). Conclusion: We construct and validate an effective CXR interpretation system based on natural language processing.

[Design and application of regional PACS based on centralized storage model].

Through integration of medical imaging resources of 12 community hospitals in the region, a community hospital diagnostic image center was established in 6th hospital. This paper describes the design of the structure and mode of data storage of regional PACS system and the establishment of a database of medical imaging, through which the medical imaging diagnostic level in the community was improved significantly. Through the application of regional PACS system, medical resource was saved and patient treatment facilitated.

Hand-Schüller-Christian disease and Erdheim-Chester disease: coexistence and discrepancy.

Langerhans cell histiocytosis (LCH) and Erdheim-Chester disease (ECD) share similar clinical features and mechanisms. In very rare circumstances, the two diseases coexist in the same patient. Here we report such a patient, who was first diagnosed with Hand-Schüller-Christian disease (HSC), a type of LCH. Several years later, the patient presented with severe exophthalmos and osteosclerosis on radiograph. New biopsy revealed ECD. We also analyze 54 cases of LCH and 6 cases of ECD diagnosed in our hospital, as well as their progression during a follow-up period of 8 years. In five cases of HSC (9.3% of LCH), a triad of central diabetes insipidus, hyperprolactinemia, and pituitary stalk thickening on magnetic resonance imaging (MRI) preceded the typical bone lesions by 4-9 years. In addition, LCH was featured as elevated plasma alkaline phosphatase (ALP), which was normal in ECD. Combined with a literature review, several features are summarized to differentiate ECD from HSC. In patients with diabetes insipidus, concomitant hyperprolactinemia and pituitary stalk thickening on MRI indicate a possible HSC. Additionally, if osteosclerosis is observed in a patient with LCH, the coexistence of ECD should be considered.

[Establishment and application of DR automatic system based on DICOM3.0 BPE].

This paper introduces a new application of PACS in our hospital. Through the integration of PACS, HIS and RIS, digital transformation is made in every step. The functional modules of Body Parts Examined in DICOM is set and good link between PACS and DR is made. So the equipment can retrieval the inspection area automatically and make adjustment on the parameters correspondingly. It makes the workflow optimized and improves the efficiency greatly.

Plating osteosynthesis of mid-distal humeral shaft fractures: minimally invasive versus conventional open reduction technique.

Results of two methods, conventional open reduction-internal plating and minimally invasive plating osteosynthesis (MIPO), in the treatment of mid-distal humeral shaft fractures were compared. Thirty-three patients were retrospectively analysed and divided into two groups. Group A (n = 17) patients were treated by MIPO and group B (n = 16) by conventional plating. The mean operation time in group A was 92.35 +/- 57.68 minutes and 103.12 +/- 31.08 minutes in group B (P = 0.513). Iatrogenic radial nerve palsy in group A was 0% (0/17) and 31.3% in group B (5/16 (P = 0.012). The mean fracture union time in group A was 15.29 +/- 4.01 weeks (range 8-24 weeks), and 21.25 +/- 13.67 weeks (range 10-58 weeks) in group B (P = 0.095). The mean UCLA end-result score in group A was 34.76 +/- 0.56 points (range 33-35), and 34.38 +/- 1.41 points (range 30-35) in group B (P = 0.299). The mean MEPI in group A was 99.41 +/- 2.43 points (range 90-100) and 99.69 +/- 1.25 points (range 95-100) in group B ( P = 0.687). When compared to the conventional plating techniques, MIPO offers advantages in terms of reduced incidence of iatrogenic radial nerve palsies and accelerated fracture union and a similar functional outcome with respect to shoulder and elbow function.