Antimoniosis: Radiological Insights into a Rare Pneumoconiosis in Miners.

Aim: Pneumoconiosis describes diseases caused by the accumulation of inorganic dust particles in the lungs, leading to tissue damage. The diagnosis relies on a history of exposure and compatible radiological findings. Background: We aimed to investigate the radiological findings in individuals exposed to antimony-inert dust relative to their working periods. Objective: Fifty-six symptomatic male antimony miners were retrospectively evaluated for demographics and chest computed tomography (CT) scans. Methods: The demographic and radiological data of patients with a history of antimony mining, who presented at our pulmonary clinic between June 2017 and June 2023, were analyzed according to the duration of exposure. Results: The study included 56 male patients with a mean age of 58.5±13.02 years and a mean exposure duration of 13.63 ± 6.82 years. CT scans showed that 73.2% (n=41) had upper and middle lung zone involvement, and 55.4% (n=31) had extensive involvement. Micronodules with centriacinar ground-glass opacities were the most common finding (n=37, 66.1%), followed by nodular opacities with irregular margins (n=22, 39.3%) and solid micronodules (n=20, 35.7%). Patients with over 20 years of exposure had significantly higher rates of respiratory and cardiovascular disease (p<0.05). Increased exposure time correlated with more extensive parenchymal involvement and higher rates of calcification in mediastinal lymph nodes, solid micronodules, nodular opacities with irregular margins, honeycombing, and conglomerate mass appearance. Conclusion: Radiological findings in pneumoconiosis generally worsen with longer exposure. Given the scarcity of up-to-date information on antimony pneumoconiosis, further studies focusing on radiological findings and chemical analyses of those exposed to antimony mine dust are essential to identify related pathologies.

Evaluation of Interstitial Lung Diseases with Deep Learning Method of Two Major Computed Tomography Patterns.

BACKGROUND: Interstitial lung diseases (ILD) encompass various disorders characterized by inflammation and/or fibrosis in the lung interstitium. These conditions produce distinct patterns in High-Resolution Computed Tomography (HRCT). OBJECTIVE: We employ a deep learning method to diagnose the most commonly encountered patterns in ILD differentially. MATERIALS AND METHODS: Patients were categorized into usual interstitial pneumonia (UIP), nonspecific interstitial pneumonia (NSIP), and normal lung parenchyma groups. VGG16 and VGG19 deep learning architectures were utilized. 85% of each pattern was used as training data for the artificial intelligence model. The models were then tasked with diagnosing the patterns in the test dataset without human intervention. Accuracy rates were calculated for both models. RESULTS: 1 The success of the VGG16 model in the test phase was 95.02% accuracy. 2 Using the same data, 98.05% accuracy results were obtained in the test phase of the VGG19 model. CONCLUSION: Deep Learning models showed high accuracy in distinguishing the two most common patterns of ILD.

Evaluation of respiratory bronchiolitis nodules with maximum intensity projection images.

OBJECTIVE: Respiratory bronchiolitis is a disease associated with heavy smoking. Computed tomography in this disease often shows symmetrical and bilaterally ill-defined circumscribed centriacinar micronodular involvement in the upper-middle lobes. The maximum intensity projection method is a kind of image processing method and provides a better evaluation of nodules and vascular structures. Our study aimed to show whether maximum intensity projection images increase the diagnostic accuracy in the detection of micronodules in respiratory bronchiolitis. METHODS: Two radiologists with different experiences (first reader: 10-year radiologist with cardiothoracic radiology experience and second reader: nonspecific radiologist with 2 years of experience) reviewed images of patients whose respiratory bronchiolitis diagnosis was supported by clinical findings. The evaluation was done independently of each other. Both conventional computed tomography images and maximum intensity projection images of the same patients were examined. The detection rates on conventional computed tomography and maximum intensity projection images were then compared. RESULTS: A total of 53 patients were evaluated, of whom 48 were men and 5 were women. The first reader detected centriacinar nodules in 42 (79.2%) patients on conventional computed tomography and centriacinar nodules in all 53 (100%) patients on maximum intensity projection images. The second reader detected centriacinar nodules in 12 (22.6%) patients on conventional computed tomography images and in 48 (90.6%) patients on maximum intensity projection images. For the less experienced reader, the detection rate of micronodules in respiratory bronchiolitis in maximum intensity projection images increased statistically significantly (p<0.001). CONCLUSION: Maximum intensity projection images in respiratory bronchiolitis increase the detectability of micronodules independently of the experience of the radiologist.

Elevating healthcare through artificial intelligence: analyzing the abdominal emergencies data set (TR_ABDOMEN_RAD_EMERGENCY) at TEKNOFEST-2022.

OBJECTIVES: The artificial intelligence competition in healthcare at TEKNOFEST-2022 provided a platform to address the complex multi-class classification challenge of abdominal emergencies using computer vision techniques. This manuscript aimed to comprehensively present the methodologies for data preparation, annotation procedures, and rigorous evaluation metrics. Moreover, it was conducted to introduce a meticulously curated abdominal emergencies data set to the researchers. METHODS: The data set underwent a comprehensive central screening procedure employing diverse algorithms extracted from the e-Nabiz (Pulse) and National Teleradiology System of the Republic of Türkiye, Ministry of Health. Full anonymization of the data set was conducted. Subsequently, the data set was annotated by a group of ten experienced radiologists. The evaluation process was executed by calculating F1 scores, which were derived from the intersection over union values between the predicted bounding boxes and the corresponding ground truth (GT) bounding boxes. The establishment of baseline performance metrics involved computing the average of the highest five F1 scores. RESULTS: Observations indicated a progressive decline in F1 scores as the threshold value increased. Furthermore, it could be deduced that class 6 (abdominal aortic aneurysm/dissection) was relatively straightforward to detect compared to other classes, with class 5 (acute diverticulitis) presenting the most formidable challenge. It is noteworthy, however, that if all achieved outcomes for all classes were considered with a threshold of 0.5, the data set's complexity and associated challenges became pronounced. CONCLUSION: This data set's significance lies in its pioneering provision of labels and GT-boxes for six classes, fostering opportunities for researchers. CLINICAL RELEVANCE STATEMENT: The prompt identification and timely intervention in cases of emergent medical conditions hold paramount significance. The handling of patients' care can be augmented, while the potential for errors is minimized, particularly amidst high caseload scenarios, through the application of AI. KEY POINTS: • The data set used in artificial intelligence competition in healthcare (TEKNOFEST-2022) provides a 6-class data set of abdominal CT images consisting of a great variety of abdominal emergencies. • This data set is compiled from the National Teleradiology System data repository of emergency radiology departments of 459 hospitals. • Radiological data on abdominal emergencies is scarce in literature and this annotated competition data set can be a valuable resource for further studies and new AI models.

Evaluation of pulmonary nodules by magnetic resonance imaging sequences: which sequence will replace computed tomography?

OBJECTIVE: This study aimed to determine the role of magnetic resonance imaging in minimizing radiation exposure, especially in the follow-up of pulmonary nodules. METHODS: Patients who applied to our hospital between April 2013 and August 2018 for various reasons and had lung-mediastinal dynamic magnetic resonance imaging and thoracic computed tomography were included in the study. A total of 194 patients were included in the study, involving 84 females and 110 males. Scanning of the nodules was done retrospectively. This study was conducted by two readers: a thoracic radiologist with 15 years of experience and a nonspecific radiologist with 4 years of experience. Evaluations were made using the double-blind method. RESULTS: Of the 194 patients, 84 (43.3%) were female and 110 (56.7%) were male. For the first reader, 135 (69.5%) nodules were detected in postcontrast T1 vibe images, 130 (67%) in T2 fast spin echo, 128 (66%) in precontrast T1 vibe, and 98 (50.5%) in T2 turbo inversion recovery magnitude sequence. For the second reader, 133 (68%) nodules were detected in postcontrast T1 vibe images, 120 (61.9%) in T2 fast spin echo, 122 (62.9%) in precontrast T1 vibe, and 99 (51%) in T2 turbo inversion recovery magnitude sequence. Capability levels were examined in detecting nodules between the first and second readers, and the ratios were reached at 0.92 in T2 fast spin echo, 0.81 in postcontrast T1 vibe images, 0.93 in precontrast T1 vibe, and 0.96 in T2 turbo inversion recovery magnitude sequence. CONCLUSION: In this study of detecting pulmonary nodules by magnetic resonance imaging, which we performed with two different readers, one of whom was an experienced thoracic radiologist, both readers found the highest detection rate in the postcontrast T1 vibe sequence.

Artificial Intelligence in Healthcare Competition (TEKNOFEST-2021): Stroke Data Set.

OBJECTIVE: The artificial intelligence competition in healthcare was organized for the first time at the annual aviation, space, and technology festival (TEKNOFEST), Istanbul/Türkiye, in September 2021. In this article, the data set preparation and competition processes were explained in detail; the anonymized and annotated data set is also provided via official website for further research. MATERIALS AND METHODS: Data set recorded over the period covering 2019 and 2020 were centrally screened from the e-Pulse and Teleradiology System of the Republic of Türkiye, Ministry of Health using various codes and filtering criteria. The data set was anonymized. The data set was prepared, pooled, curated, and annotated by 7 radiologists. The training data set was shared with the teams via a dedicated file transfer protocol server, which could be accessed using private usernames and passwords given to the teams under a nondisclosure agreement signed by the representative of each team. RESULTS: The competition consisted of 2 stages. In the first stage, teams were given 192 digital imaging and communications in medicine images that belong to 1 of 3 possible categories namely, hemorrhage, ischemic, or non-stroke. Teams were asked to classify each image as either stroke present or absent. In the second stage of the competition, qualifying 36 teams were given 97 digital imaging and communications in medicine images that contained hemorrhage, ischemia, or both lesions. Among the employed methods, Unet and DeepLabv3 were the most frequently observed ones. CONCLUSION: Artificial intelligence competitions in healthcare offer good opportunities to collect data reflecting various cases and problems. Especially, annotated data set by domain experts is more valuable.