POLCOVID: a multicenter multiclass chest X-ray database (Poland, 2020-2021).

The outbreak of the SARS-CoV-2 pandemic has put healthcare systems worldwide to their limits, resulting in increased waiting time for diagnosis and required medical assistance. With chest radiographs (CXR) being one of the most common COVID-19 diagnosis methods, many artificial intelligence tools for image-based COVID-19 detection have been developed, often trained on a small number of images from COVID-19-positive patients. Thus, the need for high-quality and well-annotated CXR image databases increased. This paper introduces POLCOVID dataset, containing chest X-ray (CXR) images of patients with COVID-19 or other-type pneumonia, and healthy individuals gathered from 15 Polish hospitals. The original radiographs are accompanied by the preprocessed images limited to the lung area and the corresponding lung masks obtained with the segmentation model. Moreover, the manually created lung masks are provided for a part of POLCOVID dataset and the other four publicly available CXR image collections. POLCOVID dataset can help in pneumonia or COVID-19 diagnosis, while the set of matched images and lung masks may serve for the development of lung segmentation solutions.

Pathological changes or technical artefacts? The problem of the heterogenous databases in COVID-19 CXR image analysis.

BACKGROUND: When the COVID-19 pandemic commenced in 2020, scientists assisted medical specialists with diagnostic algorithm development. One scientific research area related to COVID-19 diagnosis was medical imaging and its potential to support molecular tests. Unfortunately, several systems reported high accuracy in development but did not fare well in clinical application. The reason was poor generalization, a long-standing issue in AI development. Researchers found many causes of this issue and decided to refer to them as confounders, meaning a set of artefacts and methodological errors associated with the method. We aim to contribute to this steed by highlighting an undiscussed confounder related to image resolution. METHODS: 20 216 chest X-ray images (CXR) from worldwide centres were analyzed. The CXRs were bijectively projected into the 2D domain by performing Uniform Manifold Approximation and Projection (UMAP) embedding on the radiomic features (rUMAP) or CNN-based neural features (nUMAP) from the pre-last layer of the pre-trained classification neural network. Additional 44 339 thorax CXRs were used for validation. The comprehensive analysis of the multimodality of the density distribution in rUMAP/nUMAP domains and its relation to the original image properties was used to identify the main confounders. RESULTS: nUMAP revealed a hidden bias of neural networks towards the image resolution, which the regular up-sampling procedure cannot compensate for. The issue appears regardless of the network architecture and is not observed in a high-resolution dataset. The impact of the resolution heterogeneity can be partially diminished by applying advanced deep-learning-based super-resolution networks. CONCLUSIONS: rUMAP and nUMAP are great tools for image homogeneity analysis and bias discovery, as demonstrated by applying them to COVID-19 image data. Nonetheless, nUMAP could be applied to any type of data for which a deep neural network could be constructed. Advanced image super-resolution solutions are needed to reduce the impact of the resolution diversity on the classification network decision.

The S-100B level, intracranial pressure, body temperature, and transcranial blood flow velocities predict the outcome of the treatment of severe brain injury.

This study evaluates the applicability of S100B levels, mean maximum velocity (Vmean) over time, pulsatility index (PI), intracranial pressure (ICP), and body temperature (T) for the prediction of the treatment of patients with traumatic brain injury (TBI). Sixty patients defined by the Glasgow Coma Scale score ≤ 8 were stratified using the Glasgow Coma Scale into 2 groups: favorable (FG: Glasgow Outcome Scale ≥ 4) and unfavorable (UG: Glasgow Outcome Scale < 4). The S100B concentration was at the time of hospital admission. Vmean was measured using transcranial Doppler. PI was derived from a transcranial Doppler examination. T was measured in the temporal artery. The differences in mean between FG and UG were tested using a bootstrap test of 10,000 repetitions with replacement. Changes in S100B, Vmean, PI, ICP, and T levels stratified by the group were calculated using the one-way aligned rank transform for nonparametric factorial analysis of variance. The reference ranges for the levels of S100B, Vmean, and PI were 0.05 to 0.23 µg/L, 30.8 to 73.17 cm/s, and 0.62 to 1.13, respectively. Both groups were defined by an increase in Vmean, a decrease in S100B, PI, and ICP levels; and a virtually constant T. The unfavorable outcome is defined by significantly higher levels of all parameters, except T. A favorable outcome is defined by S100B < 3 mg/L, PI < 2.86, ICP > 25 mm Hg, and Vmean > 40 cm/s. The relationships provided may serve as indicators of the results of the TBI treatment.

Biomechanical evaluation of single- and multi-level anterior cervical discectomy and fusion with polyetheretherketone cages: radiological and clinical outcomes.

OBJECTIVE: The aim of this study was to analyse the outcomes of single- and multi-level anterior cervical discectomy and fusion (ACDF) with standalone polyetheretherketone (PEEK) cages, with particular emphasis on the risk of secondary adjacent segment disease. MATERIALS AND METHODS: This retrospective study included 30 patients with single- or multi-level cervical disc herniation. Before the ACDF, and one year thereafter, the patients underwent clinical and radiological evaluation including determination of cervical pain severity with a numerical rating scale (NRS), and a survey with a Polish adaptation of the neck disability index questionnaire (NDI-PL). Biomechanical parameters of the cervical spine were determined using the Cobb method. RESULTS: One year after ACDF, all patients had achieved complete fusions, and 97% showed a significant reduction of pain severity. Also, a significant decrease in all NDI-PL indices was observed. A significant decrease in overall cervical spine mobility coexisted with a significant increase in the mobility of the segment above the one operated upon and a non-significant decrease in the mobility of the segment below. No statistically significant change was found in the intervertebral disc space height (IVH) above and below the operated segment, and no evidence of degeneration within the segments adjacent to the operated one was documented. CONCLUSION: One- and two-level ACDF with standalone PEEK cages provided high fusion rates. Surgical spondylosis contributed to a reduction of spinal mobility despite the hypermobility in adjacent spinal segments. No degeneration in adjacent spinal segments was documented within a year of ACDF, and the treatment seemed to improve patients' quality of life.