Deep learning combined with radiomics may optimize the prediction in differentiating high-grade lung adenocarcinomas in ground glass opacity lesions on CT scans

2020 ◽  
Vol 129 ◽  
pp. 109150
Author(s):  
Xing Wang ◽  
Li Zhang ◽  
Xin Yang ◽  
Lei Tang ◽  
Jie Zhao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Ground Glass Opacity ◽  
Ct Scans ◽  
Ground Glass ◽  
High Grade ◽  
Lung Adenocarcinomas

scholarly journals Studi Awal Deteksi COVID-19 Menggunakan Citra CT Berbasis Deep Learning

2020 ◽  
Vol 7 (3) ◽  
pp. 629
Author(s):  
Windra Swastika
Keyword(s):  
Deep Learning ◽  
Image Data ◽  
Ground Glass Opacity ◽  
Normal Subjects ◽  
Ct Images ◽  
Ground Glass ◽  
Virus Spread ◽  
Training Process ◽  
Specificity And Sensitivity ◽  
Preliminary Study

<p class="Abstrak">Pada bulan Desember 2019, virus COVID-19 menyebar ke banyak negara, termasuk di Indonesia yang kemudian menjadi pandemi dan menimbulkan masalah serius karena masih belum adanya vaksin untuk mencegah penularan. Uji spesimen saluran nafas atas dan saluran nafas bawah saat ini merupakan salah satu metode yang efektif untuk mengetahui apakah seseorang terinfeksi COVID-19 atau tidak. Salah satu indikasi dari infeksi COVID-19 adalah sesak nafas atau pneumonia serta munculnya <em>ground-glass opacity</em> pada citra CT. Penelitian ini merupakan studi awal untuk melihat apakah citra CT dari organ thorax dapat digunakan sebagai alternatif untuk mendeteksi infeksi virus COVID-19. Deep learning digunakan untuk membuat sebuah model dengan citra CT sebagai masukan. Total 140 data citra CT yang terbagi menjadi 2 yaitu citra dari pasien terinfeksi dan citra dari subjek normal digunakan sebagai masukan pada deep learning. Proses pelatihan dilakukan menggunakan CNN dengan arsitektur VGG16 dan optimizer SGD dan Adam. Hasil yang didapatkan adalah akurasi sebesar 92,86% untuk mengklasifikasikan infeksi COVID-19 dan normal. Nilai spesifisitas dan sensitivitas sebesar 100% dan 85,71% untuk pelatihan dengan menggunakan optimizer SGD.</p><p class="Abstrak"> </p><p class="Abstrak"><em><strong>Abstract</strong></em></p><p class="Abstract"><em>In December 2019, the COVID-19 virus spread to many countries, including Indonesia which later became a pandemic and caused serious problems because there was still no vaccine to prevent transmission. Tests of upper and lower respiratory tract specimens are now an effective method of finding whether a person is infected with COVID-19 or not. One indication of COVID-19 infection is shortness of breath or pneumonia and the appearance of ground-glass opacity on CT images. This research is a preliminary study to see whether CT images of the thorax organs can be used as an alternative to detect COVID-19 virus. The deep learning is used to create a model with CT images as input. A total of 140 CT image data which are divided into 2 images from infected patients and images from normal subjects are used as input for deep learning. The training process is carried out using CNN with VGG16 architecture and SGD and Adam optimizers. The results obtained are 92.86% accuracy for classifying COVID-19 infections and normal. Specificity and sensitivity values were 100% and 85.71% for training using the SGD optimizer.</em></p><p class="Abstrak"><em><strong><br /></strong></em></p>

scholarly journals CT diagnostics and monitoring of the course of viral pneumonia caused by the SARS-CoV-2 virus during the work of the “COVID-19 Hospital”, based on the Federal Specialized Medical Scientific Center

2020 ◽  
Vol 24 (2) ◽  
pp. 11-36 ◽  
Author(s):  
G. G. Kаrmаzаnovsky ◽  
K. A. Zamyatina ◽  
V. I. Stashkiv ◽  
M. Yu. Shantarevich ◽  
E. V. Kondratyev ◽  
...  
Keyword(s):  
Lung Tissue ◽  
Ground Glass Opacity ◽  
Ct Scans ◽  
Ground Glass ◽  
Initial Examination ◽  
Ct Examination ◽  
Number Of Patients ◽  
Hospitalization Period ◽  
Pcr Test

Purpose. The research goal comprises primary analysis of CT examinations results and their interpretation by comparing with the data already available in the literature.Material and methods. During the period from April 17, 2020 to May 18, 2020, 830 chest CT scans were performed and results of 123 CDs with CT scans made by other institutions were interpreted. Follow-up examinations were carried out every 3–4 days or when clinical presentation changed. At the primary stage, we have analysed in a more detail way a group of 69 patients, who were diagnosed with CT-3 or CT-4 volume of lung damage at least once during hospitalization. The patients underwent PCR analysis three times during hospitalization. Among 69 patients, 34 patients had a positive PCR test at least once, the remaining 35 patients had a clinic, corresponding with this disease.Results. At the initial examination, ground-glass opacity prevailed, as it was observed in 44 cases (64%), and lung tissue consolidation was observed in 25 cases (36%) in a group of 69 patients. When comparing the two groups, the average age of the patients with consolidation changes was statistically significantly lower than one of the group where ground-glass opacity prevailed – 51.7 and 59.4 years, respectively (p = 0.01) In the group of patients with pulmonary tissue consolidation, there were fewer concomitant diseases, fatal outcomes, positive PCR test results, a shorter hospitalization period, and fewer cases of tocilizumab administration were noted. At the initial examination the average percentage of pulmonary parenchyma involvement in the group of patients with lung tissue consolidation was higher (63.3%; p = 0.04), follow-up examinations showed c statistically significantly lower average values of the increase in the percentage of involvement of the parenchyma, which acquired negative values after the third CT scan (8.3 after the 2nd CT and −5.2 after the 3rd CT versus 18.5 and 3 in the GGO glass group; p = 0.02 and 0.03, respectively). No visible differences in CT between the period from the onset of the disease and the predominant symptom in CT were revealed. Meanwhile, on the 5th day (the day of the check-up CT examination) the largest number of patients was determined in both groups.Conclusion. An analysis of our experience during the first month of operation of Covid-19 Hospital is presented. According to our data, the appearance of consolidation at the initial CT examination is probably not related to the period, when the disease has been in progress, and may be associated with a more favorable course of the process.

scholarly journals Deep-learning algorithms for the interpretation of chest radiographs to aid in the triage of COVID-19 patients: A multicenter retrospective study

PLoS ONE ◽  
2020 ◽  
Vol 15 (11) ◽  
pp. e0242759
Author(s):  
Se Bum Jang ◽  
Suk Hee Lee ◽  
Dong Eun Lee ◽  
Sin-Youl Park ◽  
Jong Kun Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Retrospective Study ◽  
Image Interpretation ◽  
Area Under The Curve ◽  
Ground Glass Opacity ◽  
Community Treatment ◽  
Ct Scans ◽  
Treatment Center ◽  
Automatic Interpretation ◽  
Radiology Reports

The recent medical applications of deep-learning (DL) algorithms have demonstrated their clinical efficacy in improving speed and accuracy of image interpretation. If the DL algorithm achieves a performance equivalent to that achieved by physicians in chest radiography (CR) diagnoses with Coronavirus disease 2019 (COVID-19) pneumonia, the automatic interpretation of the CR with DL algorithms can significantly reduce the burden on clinicians and radiologists in sudden surges of suspected COVID-19 patients. The aim of this study was to evaluate the efficacy of the DL algorithm for detecting COVID-19 pneumonia on CR compared with formal radiology reports. This is a retrospective study of adult patients that were diagnosed as positive COVID-19 cases based on the reverse transcription polymerase chain reaction among all the patients who were admitted to five emergency departments and one community treatment center in Korea from February 18, 2020 to May 1, 2020. The CR images were evaluated with a publicly available DL algorithm. For reference, CR images without chest computed tomography (CT) scans classified as positive for COVID-19 pneumonia were used given that the radiologist identified ground-glass opacity, consolidation, or other infiltration in retrospectively reviewed CR images. Patients with evidence of pneumonia on chest CT scans were also classified as COVID-19 pneumonia positive outcomes. The overall sensitivity and specificity of the DL algorithm for detecting COVID-19 pneumonia on CR were 95.6%, and 88.7%, respectively. The area under the curve value of the DL algorithm for the detection of COVID-19 with pneumonia was 0.921. The DL algorithm demonstrated a satisfactory diagnostic performance comparable with that of formal radiology reports in the CR-based diagnosis of pneumonia in COVID-19 patients. The DL algorithm may offer fast and reliable examinations that can facilitate patient screening and isolation decisions, which can reduce the medical staff workload during COVID-19 pandemic situations.

scholarly journals Detection of pulmonary ground-glass opacity based on deep learning computer artificial intelligence

2019 ◽  
Vol 18 (1) ◽  
Author(s):  
Wenjing Ye ◽  
Wen Gu ◽  
Xuejun Guo ◽  
Ping Yi ◽  
Yishuang Meng ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Deep Learning ◽  
Ground Glass Opacity ◽  
Ground Glass

scholarly journals Domain-Adaptive Artificial Intelligence-Based Model for Personalized Diagnosis of Trivial Lesions Related to COVID-19 in Chest Computed Tomography Scans

2021 ◽  
Vol 11 (10) ◽  
pp. 1008
Author(s):  
Muhammad Owais ◽  
Na Rae Baek ◽  
Kang Ryoung Park
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
State Of The Art ◽  
Area Under The Curve ◽  
Ground Glass Opacity ◽  
Visual Assessment ◽  
Ct Scans ◽  
Execution Speed ◽  
Ct Data ◽  
Personalized Diagnosis

Background: Early and accurate detection of COVID-19-related findings (such as well-aerated regions, ground-glass opacity, crazy paving and linear opacities, and consolidation in lung computed tomography (CT) scan) is crucial for preventive measures and treatment. However, the visual assessment of lung CT scans is a time-consuming process particularly in case of trivial lesions and requires medical specialists. Method: A recent breakthrough in deep learning methods has boosted the diagnostic capability of computer-aided diagnosis (CAD) systems and further aided health professionals in making effective diagnostic decisions. In this study, we propose a domain-adaptive CAD framework, namely the dilated aggregation-based lightweight network (DAL-Net), for effective recognition of trivial COVID-19 lesions in CT scans. Our network design achieves a fast execution speed (inference time is 43 ms on a single image) with optimal memory consumption (almost 9 MB). To evaluate the performances of the proposed and state-of-the-art models, we considered two publicly accessible datasets, namely COVID-19-CT-Seg (comprising a total of 3520 images of 20 different patients) and MosMed (including a total of 2049 images of 50 different patients). Results: Our method exhibits average area under the curve (AUC) up to 98.84%, 98.47%, and 95.51% for COVID-19-CT-Seg, MosMed, and cross-dataset, respectively, and outperforms various state-of-the-art methods. Conclusions: These results demonstrate that deep learning-based models are an effective tool for building a robust CAD solution based on CT data in response to present disaster of COVID-19.

Sign in / Sign up

Export Citation Format

Share Document