Background and motivation: Lung computed tomography (CT) techniques are high-resolution and are well adopted in the intensive care unit (ICU) for COVID-19 disease control classification. Most artificial intelligence (AI) systems do not undergo generalization and are typically overfitted. Such trained AI systems are not practical for clinical settings and therefore do not give accurate results when executed on unseen data sets. We hypothesize that ensemble deep learning (EDL) is superior to deep transfer learning (TL) in both non-augmented and augmented frameworks. Methodology: The system consists of a cascade of quality control, ResNet–UNet-based hybrid deep learning for lung segmentation, and seven models using TL-based classification followed by five types of EDL’s. To prove our hypothesis, five different kinds of data combinations (DC) were designed using a combination of two multicenter cohorts—Croatia (80 COVID) and Italy (72 COVID and 30 controls)—leading to 12,000 CT slices. As part of generalization, the system was tested on unseen data and statistically tested for reliability/stability. Results: Using the K5 (80:20) cross-validation protocol on the balanced and augmented dataset, the five DC datasets improved TL mean accuracy by 3.32%, 6.56%, 12.96%, 47.1%, and 2.78%, respectively. The five EDL systems showed improvements in accuracy of 2.12%, 5.78%, 6.72%, 32.05%, and 2.40%, thus validating our hypothesis. All statistical tests proved positive for reliability and stability. Conclusion: EDL showed superior performance to TL systems for both (a) unbalanced and unaugmented and (b) balanced and augmented datasets for both (i) seen and (ii) unseen paradigms, validating both our hypotheses.
Background: The aim of our study was to determine the normal ventricular size of the intracranial cavity based on the bicaudate index (BCI) calculation which was obtained through the computed tomography. Methods: We reviewed consecutive CT examinations in 1406 patients with symmetrical intracranial ventricular system and normal density of the white and gray matter, and without neurological deficits or psychiatric disorders. Results: Our series included 711 female and 695 male examinees; aged from 3 months to 89 years, mean age 46,3 ± 23,9 years. The positive correlation between age and bicaudate indexes was significant (p < 0.01). In our series, upper limit of the BCI (99th percentile) was 0.13 for the first three decades. In the fourth and fifth decade it increases to 0.15 and 0.19, respectively. Conclusions: After the age of 50, BCI progressively increases and rises to levels that are nearly double those found in the age group of 30 or younger. This is important in CT assessments of potential developments of hydrocephalus in older patients.
We describe a case of acute obstructive hydrocephalus as a consequence of compression of the brainstem by a large aneurysm of the basilar artery (BA) in a 62-year-old male. After the insertion of the ventriculoperitoneal shunt (VPS), we encountered the "locked-in syndrome" clinical condition. "Locked-in syndrome" is a clinical state characterized by quadriplegia and anarthria with preserved consciousness, most commonly caused by ischemia in the ventral part of pons.