ABSTRACT
Hepatocellular carcinoma (HCC) has become the 4th leading cause of cancer-related deaths, with high social, economical and health implications. Imaging techniques such as multiphase computed tomography (CT) have been successfully used for diagnosis of liver tumors such as HCC in a feasible and accurate way and its interpretation relies mainly on comparing the appearance of the lesions in the different contrast phases of the exam. Recently, some researchers have been dedicated to the development of tools based on machine learning (ML) algorithms, especially by deep learning techniques, to improve the diagnosis of liver lesions in imaging exams. However, the lack of standardization in the naming of the CT contrast phases in the DICOM metadata is a problem for real-life deployment of machine learning tools. Therefore, it is important to correctly identify the exam phase based only on the image and not on the exam metadata, which is unreliable. Motivated by this problem, we successfully created an annotation platform and implemented a convolutional neural network (CNN) to automatically identify the CT scan phases in the HCFMUSP database in the city of São Paulo, Brazil. We improved this algorithm with hyperparameter tuning and evaluated it with cross validation methods. Comparing its predictions with the radiologists annotation, it achieved an accuracy of 94.6%, 98% and 100% in the testing dataset for the slice, volume and exam evaluation, respectively.
Subject(s)
Carcinoma, Hepatocellular , Deep Learning , Liver Neoplasms , Humans , Carcinoma, Hepatocellular/diagnostic imaging , Liver Neoplasms/diagnostic imaging , Brazil , Tomography, X-Ray Computed/methods , ComputersABSTRACT
OBJECTIVE: To investigate the relationship between lung lesion burden (LLB) found on chest computed tomography (CT) and 30-day mortality in hospitalized patients with high clinical suspicion of coronavirus disease 2019 (COVID-19), accounting for tomographic dynamic changes. METHODS: Patients hospitalized with high clinical suspicion of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in a dedicated and reference hospital for COVID-19, having undergone at least one RT-PCR test, regardless of the result, and with one CT compatible with COVID-19, were retrospectively studied. Clinical and laboratory data upon admission were assessed, and LLB found on CT was semi-quantitatively evaluated through visual analysis. The primary outcome was 30-day mortality after admission. Secondary outcomes, including the intensive care unit (ICU) admission, mechanical ventilation used, and length of stay (LOS), were assessed. RESULTS: A total of 457 patients with a mean age of 57±15 years were included. Among these, 58% presented with positive RT-PCR result for COVID-19. The median time from symptom onset to RT-PCR was 8 days [interquartile range 6-11 days]. An initial LLB of ≥50% using CT was found in 201 patients (44%), which was associated with an increased crude at 30-day mortality (31% vs. 15% in patients with LLB of <50%, p<0.001). An LLB of ≥50% was also associated with an increase in the ICU admission, the need for mechanical ventilation, and a prolonged LOS after adjusting for baseline covariates and accounting for the CT findings as a time-varying covariate; hence, patients with an LLB of ≥50% remained at a higher risk at 30-day mortality (adjusted hazard ratio 2.17, 95% confidence interval 1.47-3.18, p<0.001). CONCLUSION: Even after accounting for dynamic CT changes in patients with both clinical and imaging findings consistent with COVID-19, an LLB of ≥50% might be associated with a higher risk of mortality.
Subject(s)
COVID-19 , Adult , Aged , Humans , Lung/diagnostic imaging , Middle Aged , Prognosis , Retrospective Studies , SARS-CoV-2 , Tomography, X-Ray ComputedABSTRACT
OBJECTIVE: To evaluate the role of chest computed tomography in patients with COVID-19 who presented initial negative result in reverse transcriptase-polymerase chain reaction (RT-PCR). METHODS: A single-center, retrospective study that evaluated 39 patients with negative RT-PCR for COVID-19, who underwent chest computed tomography and had a final clinical or serological diagnosis of COVID-19. The visual tomographic classification was evaluated according to the Consensus of the Radiological Society of North America and software developed with artificial intelligence for automatic detection of findings and chance estimation of COVID-19. RESULTS: In the visual tomographic analysis, only one of them (3%) presented computed tomography classified as negative, 69% were classified as typical and 28% as indeterminate. In the evaluation using the software, only four (about 10%) had a probability of COVID-19 <25%. CONCLUSION: Computed tomography can play an important role in management of suspected cases of COVID-19 with initial negative results in RT-PCR, especially considering those patients outside the ideal window for sample collection for RT-PCR.
Subject(s)
COVID-19 , Artificial Intelligence , Humans , Lung , Retrospective Studies , Reverse Transcriptase Polymerase Chain Reaction , SARS-CoV-2 , Tomography, X-Ray ComputedABSTRACT
OBJECTIVE: To determine the correlation between the two tomographic classifications for coronavirus disease (COVID-19), COVID-19 Reporting and Data System (CORADS) and Radiological Society of North America Expert Consensus Statement on Reporting Chest Computed Tomography (CT) Findings Related to COVID-19 (RSNA), in the Brazilian population and to assess the agreement between reviewers with different experience levels. METHODS: Chest CT images of patients with reverse transcriptase-polymerase chain reaction (RT-PCR)-positive COVID-19 were categorized according to the CORADS and RSNA classifications by radiologists with different levels of experience and who were initially unaware of the RT-PCR results. The inter- and intra-observer concordances for each of the classifications were calculated, as were the concordances between classifications. RESULTS: A total of 100 patients were included in this study. The RSNA classification showed an almost perfect inter-observer agreement between reviewers with similar experience levels, with a kappa coefficient of 0.892 (95% confidence interval [CI], 0.788-0.995). CORADS showed substantial agreement among reviewers with similar experience levels, with a kappa coefficient of 0.642 (95% CI, 0.491-0.793). There was inter-observer variation when comparing less experienced reviewers with more experienced reviewers, with the highest kappa coefficient of 0.396 (95% CI, 0.255-0.588). There was a significant correlation between both classifications, with a Kendall coefficient of 0.899 (p<0.001) and substantial intra-observer agreement for both classifications. CONCLUSION: The RSNA and CORADS classifications showed excellent inter-observer agreement for reviewers with the same level of experience, although the agreement between less experience reviewers and the reviewer with the most experience was only reasonable. Combined analysis of both classifications with the first RT-PCR results did not reveal any false-negative results for detecting COVID-19 in patients.
Subject(s)
COVID-19 , Coronavirus , Brazil , Humans , Observer Variation , SARS-CoV-2 , Tomography, X-Ray ComputedABSTRACT
OBJECTIVE: To investigate the relationship between lung lesion burden (LLB) found on chest computed tomography (CT) and 30-day mortality in hospitalized patients with high clinical suspicion of coronavirus disease 2019 (COVID-19), accounting for tomographic dynamic changes. METHODS: Patients hospitalized with high clinical suspicion of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in a dedicated and reference hospital for COVID-19, having undergone at least one RT-PCR test, regardless of the result, and with one CT compatible with COVID-19, were retrospectively studied. Clinical and laboratory data upon admission were assessed, and LLB found on CT was semi-quantitatively evaluated through visual analysis. The primary outcome was 30-day mortality after admission. Secondary outcomes, including the intensive care unit (ICU) admission, mechanical ventilation used, and length of stay (LOS), were assessed. RESULTS: A total of 457 patients with a mean age of 57±15 years were included. Among these, 58% presented with positive RT-PCR result for COVID-19. The median time from symptom onset to RT-PCR was 8 days [interquartile range 6-11 days]. An initial LLB of ≥50% using CT was found in 201 patients (44%), which was associated with an increased crude at 30-day mortality (31% vs. 15% in patients with LLB of <50%, p<0.001). An LLB of ≥50% was also associated with an increase in the ICU admission, the need for mechanical ventilation, and a prolonged LOS after adjusting for baseline covariates and accounting for the CT findings as a time-varying covariate; hence, patients with an LLB of ≥50% remained at a higher risk at 30-day mortality (adjusted hazard ratio 2.17, 95% confidence interval 1.47-3.18, p<0.001). CONCLUSION: Even after accounting for dynamic CT changes in patients with both clinical and imaging findings consistent with COVID-19, an LLB of ≥50% might be associated with a higher risk of mortality.
Subject(s)
Humans , Adult , Middle Aged , Aged , COVID-19 , Prognosis , Tomography, X-Ray Computed , Retrospective Studies , SARS-CoV-2 , Lung/diagnostic imagingABSTRACT
ABSTRACT Objective To evaluate the role of chest computed tomography in patients with COVID-19 who presented initial negative result in reverse transcriptase-polymerase chain reaction (RT-PCR). Methods A single-center, retrospective study that evaluated 39 patients with negative RT-PCR for COVID-19, who underwent chest computed tomography and had a final clinical or serological diagnosis of COVID-19. The visual tomographic classification was evaluated according to the Consensus of the Radiological Society of North America and software developed with artificial intelligence for automatic detection of findings and chance estimation of COVID-19. Results In the visual tomographic analysis, only one of them (3%) presented computed tomography classified as negative, 69% were classified as typical and 28% as indeterminate. In the evaluation using the software, only four (about 10%) had a probability of COVID-19 <25%. Conclusion Computed tomography can play an important role in management of suspected cases of COVID-19 with initial negative results in RT-PCR, especially considering those patients outside the ideal window for sample collection for RT-PCR.
RESUMO Objetivo Avaliar o papel da tomografia computadorizada de tórax em pacientes com COVID-19 que apresentaram reação em cadeia da polimerase via transcriptase reversa (RT-PCR) inicial falsamente negativa. Métodos Estudo retrospectivo de centro único que avaliou 39 pacientes com RT-PCR negativa para COVID-19, submetidos à tomografia computadorizada de tórax e que tiveram diagnóstico final clínico ou serológico de COVID-19. A classificação tomográfica visual foi avaliada de acordo com o Consenso da Radiological Society of North America e o software desenvolvido com inteligência artificial para detecção automática de achados e estimativa de probabilidade de COVID-19. Resultados Na análise tomográfica visual, somente um deles (3%) apresentou tomografia computadorizada classificada como tendo resultado negativo, 69% foram classificados como típicos e 28% como indeterminados. Na avaliação com uso de software, somente quatro (cerca de 10%) tiveram probabilidade de COVID-19 <25%. Conclusão A tomografia computadorizada pode desempenhar papel importante no manejo de casos suspeitos de COVID-19 com RT-PCR inicialmente negativa, principalmente levando-se em consideração os pacientes que estão fora da janela ideal para coleta de amostra para RT-PCR.
Subject(s)
Humans , COVID-19 , Artificial Intelligence , Tomography, X-Ray Computed , Retrospective Studies , Reverse Transcriptase Polymerase Chain Reaction , SARS-CoV-2 , LungABSTRACT
OBJECTIVE: To determine the correlation between the two tomographic classifications for coronavirus disease (COVID-19), COVID-19 Reporting and Data System (CORADS) and Radiological Society of North America Expert Consensus Statement on Reporting Chest Computed Tomography (CT) Findings Related to COVID-19 (RSNA), in the Brazilian population and to assess the agreement between reviewers with different experience levels. METHODS: Chest CT images of patients with reverse transcriptase-polymerase chain reaction (RT-PCR)-positive COVID-19 were categorized according to the CORADS and RSNA classifications by radiologists with different levels of experience and who were initially unaware of the RT-PCR results. The inter- and intra-observer concordances for each of the classifications were calculated, as were the concordances between classifications. RESULTS: A total of 100 patients were included in this study. The RSNA classification showed an almost perfect inter-observer agreement between reviewers with similar experience levels, with a kappa coefficient of 0.892 (95% confidence interval [CI], 0.788-0.995). CORADS showed substantial agreement among reviewers with similar experience levels, with a kappa coefficient of 0.642 (95% CI, 0.491-0.793). There was inter-observer variation when comparing less experienced reviewers with more experienced reviewers, with the highest kappa coefficient of 0.396 (95% CI, 0.255-0.588). There was a significant correlation between both classifications, with a Kendall coefficient of 0.899 (p<0.001) and substantial intra-observer agreement for both classifications. CONCLUSION: The RSNA and CORADS classifications showed excellent inter-observer agreement for reviewers with the same level of experience, although the agreement between less experience reviewers and the reviewer with the most experience was only reasonable. Combined analysis of both classifications with the first RT-PCR results did not reveal any false-negative results for detecting COVID-19 in patients.
Subject(s)
Humans , Coronavirus Infections , Coronavirus , Brazil , Tomography, X-Ray Computed , Observer Variation , BetacoronavirusABSTRACT
Soft-tissue calcifications are extremely common. Because the imaging findings are nonspecific, soft-tissue calcifications are often problematic for radiologists, sometimes prompting unnecessary interventions. In addition, the nomenclature is quite confusing. Classically, soft-tissue calcifications are divided into four categories, by mechanism of formation-dystrophic, iatrogenic, metastatic, and idiopathic-depending on the clinical and biochemical correlation. However, it is also possible to classify such calcifications by compartment, and that classification can be quite useful in the radiological diagnostic assessment. In this article, we illustrate the main causes of soft-tissue calcifications, organizing them according to their anatomical and pathophysiological aspects, thus narrowing the differential diagnosis.
Calcificações de partes moles são achados extremamente comuns e inespecíficos nos exames de imagem e, por isso, frequentemente são fonte de confusão por parte dos radiologistas, desencadeando, por vezes, intervenções desnecessárias. Além disso, a nomenclatura atribuída é muito confusa. Classicamente, dividem-se as calcificações de partes moles, conforme seu mecanismo de formação, em calcificações distróficas, iatrogênicas, metastáticas e idiopáticas, dependendo de correlação clinicolaboratorial, porém, também é possível uma classificação compartimental das calcificações, que pode ser muito útil na propedêutica radiológica. Neste trabalho, ilustramos didaticamente as principais causas de calcificações de partes moles organizando-as de acordo com aspectos anatômicos e fisiopatológicos, estreitando os diagnósticos diferenciais.