A machine learning tool to improve prediction of mediastinal lymph node metastases in non-small cell lung cancer using routinely obtainable [18F]FDG-PET/CT parameters.

BACKGROUND: In patients with non-small cell lung cancer (NSCLC), accuracy of [18F]FDG-PET/CT for pretherapeutic lymph node (LN) staging is limited by false positive findings. Our aim was to evaluate machine learning with routinely obtainable variables to improve accuracy over standard visual image assessment. METHODS: Monocentric retrospective analysis of pretherapeutic [18F]FDG-PET/CT in 491 consecutive patients with NSCLC using an analog PET/CT scanner (training + test cohort, n = 385) or digital scanner (validation, n = 106). Forty clinical variables, tumor characteristics, and image variables (e.g., primary tumor and LN SUVmax and size) were collected. Different combinations of machine learning methods for feature selection and classification of N0/1 vs. N2/3 disease were compared. Ten-fold nested cross-validation was used to derive the mean area under the ROC curve of the ten test folds ("test AUC") and AUC in the validation cohort. Reference standard was the final N stage from interdisciplinary consensus (histological results for N2/3 LNs in 96%). RESULTS: N2/3 disease was present in 190 patients (39%; training + test, 37%; validation, 46%; p = 0.09). A gradient boosting classifier (GBM) with 10 features was selected as the final model based on test AUC of 0.91 (95% confidence interval, 0.87-0.94). Validation AUC was 0.94 (0.89-0.98). At a target sensitivity of approx. 90%, test/validation accuracy of the GBM was 0.78/0.87. This was significantly higher than the accuracy based on "mediastinal LN uptake > mediastinum" (0.7/0.75; each p < 0.05) or combined PET/CT criteria (PET positive and/or LN short axis diameter > 10 mm; 0.68/0.75; each p < 0.001). Harmonization of PET images between the two scanners affected SUVmax and visual assessment of the LNs but did not diminish the AUC of the GBM. CONCLUSIONS: A machine learning model based on routinely available variables from [18F]FDG-PET/CT improved accuracy in mediastinal LN staging compared to established visual assessment criteria. A web application implementing this model was made available.

Causes of death and comorbidities in hospitalized patients with COVID-19.

Infection by the new corona virus strain SARS-CoV-2 and its related syndrome COVID-19 has been associated with more than two million deaths worldwide. Patients of higher age and with preexisting chronic health conditions are at an increased risk of fatal disease outcome. However, detailed information on causes of death and the contribution of pre-existing health conditions to death yet is missing, which can be reliably established by autopsy only. We performed full body autopsies on 26 patients that had died after SARS-CoV-2 infection and COVID-19 at the Charité University Hospital Berlin, Germany, or at associated teaching hospitals. We systematically evaluated causes of death and pre-existing health conditions. Additionally, clinical records and death certificates were evaluated. We report findings on causes of death and comorbidities of 26 decedents that had clinically presented with severe COVID-19. We found that septic shock and multi organ failure was the most common immediate cause of death, often due to suppurative pulmonary infection. Respiratory failure due to diffuse alveolar damage presented as immediate cause of death in fewer cases. Several comorbidities, such as hypertension, ischemic heart disease, and obesity were present in the vast majority of patients. Our findings reveal that causes of death were directly related to COVID-19 in the majority of decedents, while they appear not to be an immediate result of preexisting health conditions and comorbidities. We therefore suggest that the majority of patients had died of COVID-19 with only contributory implications of preexisting health conditions to the mechanism of death.

A diagnostic predicament: activated sarcoidosis or pulmonary histoplasmosis. A case report.

We report a case of a 41-year-old man presenting with persisting fevers over 2 weeks. The patient had spent 4 weeks in Central America. He was in control of a stable stage II sarcoidosis. Laboratory and various microbiological tests as well as chest radiography led to no diagnosis. Activated sarcoidosis was hypothesized as the most likely diagnosis. However, we considered an infectious process as a differential diagnosis, in detail, the travel history imposed histoplasmosis. Chest-CT documented localized interstitial consolidations. Bronchoscopy with bronchoalveolar lavage (BAL) and biopsy was performed. Results of BAL fluid, biopsy, distinct sarcoidosis serum markers and a borderline positive histoplasmosis-serology yielded in a diagnostic dilemma as no distinct diagnosis was drawable. After the patient was already started on a prednisolone trial, the final diagnosis - pulmonary histoplasmosis - could be achieved via positive culture and PCR out of the BAL fluid. This case shows the difficult differentiation between an acute exacerbation of a chronic pulmonary disease and a concomitant infection, which was especially aggravated in this case as the histoplasmosis masqueraded an acute picture of sarcoidosis.