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.

Noticable Findings in Cranial MRI of the Patients with Idiopathic Intracranial Hypertension.

OBJECTIVES: Idiopathic intracranial hypertension (IIH) is characterized by increased intracranial pressure. We investigated the optic nerve, Meckel's cavity, internal carotid artery (ICA) and pituitary findings of IIH by Cranial Magnetic Resonance Imaging (MRI). METHODS: Cranial MRI images of 35 adult patients with IIH and 35 adult subjects with normal cranial MRI results (control) were evaluated. Optic nerve diameter (OND), optic nerve sheat diameter (ONSD), OND/ONSD ratio, ON tortuosity, ON protrusion, posterior scleral flattening, ICA transverse diameter, CSF distance in Meckel's cavity and ICA transverse diameter/CSF distance in Meckel's cavity ratio, and pituitary gland measurements (height and transverse dimension; and Optic chiasm- pituitary gland distance) were measured. RESULTS: OND and ONSD of the IIH group were significantly higher than those of the control groups at anterior and posterior measurements (p < 0.05). OND/ONSD ratio of the IIH group was lower at anterior measurement; and higher at the posterior measurement than the control group (p < 0.05). Right ICA transverse diameter and bilateral CSF distance in Meckel's cavity of the IIH group were higher than those of the control. Optic chiasm- pituitary gland distance of the IIH group was significantly higher than that of the control group (p < 0.05). CONCLUSION: OND/ONSD ratio is different in anterior and posterior measurements. So we recommend measuring OND and ONSD separately in IIH patients to use in the clinical practice. Similarly, optic chiasm-pituitary gland distance is also another point to note for IIH patients on MRI. Our paper adds new approach to IIH in terms of OND/ONSD ratio.

The diagnostic value of neurogranin in patients with carbon monoxide poisoning: Can it show early neurological damage?

BACKGROUND AND AIM: Carbon monoxide poisoning is a toxicological emergency that causes neurological complications. High serum neurogranin can be detected in acute or chronic conditions where brain tissue is damaged. This study aimed to investigate the diagnostic value of serum neurogranin level and its role in demonstrating neurological damage in patients admitted to the emergency department with carbon monoxide poisoning. MATERIALS AND METHODS: The study was conducted prospectively on patients with carbon monoxide poisoning (patient group) and healthy volunteers (control group). Demographic characteristics and serum neurogranin level of all participants and symptoms at admission, neurological examination findings, laboratory results, and Diffusion-Weighted Magnetic Resonance Imaging results of the patient group were recorded. We used an independent sample t-test to compare neurogranin levels and bivariate correlation analysis to compare the relationship between serum neurogranin levels and data belonging to the patient group. RESULTS: Sixty eight participants (patient group, n = 36; control group, n = 32) were included in the study. Serum neurogranin level was significantly higher in patients with carbon monoxide poisoning (0.31 ± 0.16 ng/ml) compared to control group (0.22 ± 0.10 ng/ml) (p = 0.015). The mean Glasgow Coma Scale of the patients with carbon monoxide poisoning was 14.59 ± 0.23, and of Diffusion Weighted Magnetic Resonance Imaging results were completely normal in 94.4% (n = 34). There was no correlation between serum neurogranin level and Diffusion Weighted Magnetic Resonance Imaging results (r = -0.011; p = 0.953). CONCLUSION: Serum neurogranin level may be a new diagnostic biomarker in patients admitted to the emergency department with carbon monoxide poisoning. The high serum neurogranin levels detected in patients with normal diffusion-weighted imaging after carbon monoxide poisoning suggest that there is neurological damage in these patients, even if imaging methods cannot detect it.

Is there a relationship between basilar artery tortuosity and vertigo?

OBJECTIVE: The aim of this study was to investigate the relationship between basilar artery (BA) tortuosity, hypogenesis/agenesis of the vertebral artery (VA), and vertigo, with the use of magnetic resonance imaging (MRI). PATIENTS AND METHODS: This case-control study included patients admitted to the outpatient clinics, who were aged 18-80 years, without any known systemic diseases. All patients were evaluated with a 1.5-tesla MRI system. BA Tortuosity, VA agenesis, and VA asymmetry were noted. BA diameter (central) and length (longitudinal) were measured. RESULTS: A total of 154 vertigo patients (46 M, 108 F; mean age of 48.95 ± 17.3 years) and 346 control subjects (112 M, 234 F; mean age of 45.12 ± 17.0 years) were included. The mean age of the vertigo patients was significantly higher than that of the control group (48.95 vs 45.12 years) (p = 0.021). The rate of BA tortuosity was higher in patients with vertigo (p = 0.030). When the participants were divided into two groups according to median age (<45 vs. ≥45 years) there was no statistically significant difference between the groups in terms of VA asymmetry (p = 0.070) and hypogenesis/agenesis (p = 0.577). There was a statistically significant difference between the groups in respect of BA tortuosity (p = 0.033), BA diameter (p < 0.001), and BA length (p < 0.001). When the study populations were divided into two groups according to the presence of vascular tortuosity, the mean age, BA diameter, and BA length values were higher in the tortuosity (+) group (all p < 0.001). CONCLUSION: These results demonstrated that vertigo and BA tortuosity rates seem to increase with age. Likewise, BA diameter and length increased with age, although there was no significant relationship with vertigo. Patients with tortuosity were significantly older, and had higher rates of VA asymmetry/agenesis, and increased BA diameter compared to subjects without tortuosity.