1
|
Aromiwura AA, Settle T, Umer M, Joshi J, Shotwell M, Mattumpuram J, Vorla M, Sztukowska M, Contractor S, Amini A, Kalra DK. Artificial intelligence in cardiac computed tomography. Prog Cardiovasc Dis 2023; 81:54-77. [PMID: 37689230 DOI: 10.1016/j.pcad.2023.09.001] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/04/2023] [Accepted: 09/04/2023] [Indexed: 09/11/2023]
Abstract
Artificial Intelligence (AI) is a broad discipline of computer science and engineering. Modern application of AI encompasses intelligent models and algorithms for automated data analysis and processing, data generation, and prediction with applications in visual perception, speech understanding, and language translation. AI in healthcare uses machine learning (ML) and other predictive analytical techniques to help sort through vast amounts of data and generate outputs that aid in diagnosis, clinical decision support, workflow automation, and prognostication. Coronary computed tomography angiography (CCTA) is an ideal union for these applications due to vast amounts of data generation and analysis during cardiac segmentation, coronary calcium scoring, plaque quantification, adipose tissue quantification, peri-operative planning, fractional flow reserve quantification, and cardiac event prediction. In the past 5 years, there has been an exponential increase in the number of studies exploring the use of AI for cardiac computed tomography (CT) image acquisition, de-noising, analysis, and prognosis. Beyond image processing, AI has also been applied to improve the imaging workflow in areas such as patient scheduling, urgent result notification, report generation, and report communication. In this review, we discuss algorithms applicable to AI and radiomic analysis; we then present a summary of current and emerging clinical applications of AI in cardiac CT. We conclude with AI's advantages and limitations in this new field.
Collapse
Affiliation(s)
| | - Tyler Settle
- Medical Imaging Laboratory, Department of Electrical and Computer Engineering, University of Louisville, Louisville, KY, USA
| | - Muhammad Umer
- Division of Cardiology, Department of Medicine, University of Louisville, Louisville, KY, USA
| | - Jonathan Joshi
- Center for Artificial Intelligence in Radiological Sciences (CAIRS), Department of Radiology, University of Louisville, Louisville, KY, USA
| | - Matthew Shotwell
- Division of Cardiology, Department of Medicine, University of Louisville, Louisville, KY, USA
| | - Jishanth Mattumpuram
- Division of Cardiology, Department of Medicine, University of Louisville, Louisville, KY, USA
| | - Mounica Vorla
- Division of Cardiology, Department of Medicine, University of Louisville, Louisville, KY, USA
| | - Maryta Sztukowska
- Clinical Trials Unit, University of Louisville, Louisville, KY, USA; University of Information Technology and Management, Rzeszow, Poland
| | - Sohail Contractor
- Center for Artificial Intelligence in Radiological Sciences (CAIRS), Department of Radiology, University of Louisville, Louisville, KY, USA
| | - Amir Amini
- Medical Imaging Laboratory, Department of Electrical and Computer Engineering, University of Louisville, Louisville, KY, USA; Center for Artificial Intelligence in Radiological Sciences (CAIRS), Department of Radiology, University of Louisville, Louisville, KY, USA
| | - Dinesh K Kalra
- Division of Cardiology, Department of Medicine, University of Louisville, Louisville, KY, USA; Center for Artificial Intelligence in Radiological Sciences (CAIRS), Department of Radiology, University of Louisville, Louisville, KY, USA.
| |
Collapse
|
2
|
Valentini A, Franchi P, Cicchetti G, Messana G, Chiffi G, Strappa C, Calandriello L, Del Ciello A, Farchione A, Preda L, Larici AR. Pulmonary Hypertension in Chronic Lung Diseases: What Role Do Radiologists Play? Diagnostics (Basel) 2023; 13:diagnostics13091607. [PMID: 37174998 PMCID: PMC10178805 DOI: 10.3390/diagnostics13091607] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/20/2023] [Revised: 04/25/2023] [Accepted: 04/26/2023] [Indexed: 05/15/2023] Open
Abstract
Pulmonary hypertension (PH) is a pathophysiological disorder, defined by a mean pulmonary arterial pressure (mPAP) > 20 mmHg at rest, as assessed by right heart catheterization (RHC). PH is not a specific disease, as it may be observed in multiple clinical conditions and may complicate a variety of thoracic diseases. Conditions associated with the risk of developing PH are categorized into five different groups, according to similar clinical presentations, pathological findings, hemodynamic characteristics, and treatment strategy. Most chronic lung diseases that may be complicated by PH belong to group 3 (interstitial lung diseases, chronic obstructive pulmonary disease, combined pulmonary fibrosis, and emphysema) and are associated with the lowest overall survival among all groups. However, some of the chronic pulmonary diseases may develop PH with unclear/multifactorial mechanisms and are included in group 5 PH (sarcoidosis, pulmonary Langerhans' cell histiocytosis, and neurofibromatosis type 1). This paper focuses on PH associated with chronic lung diseases, in which radiological imaging-particularly computed tomography (CT)-plays a crucial role in diagnosis and classification. Radiologists should become familiar with the hemodynamical, physiological, and radiological aspects of PH and chronic lung diseases in patients at risk of developing PH, whose prognosis and treatment depend on the underlying disease.
Collapse
Affiliation(s)
- Adele Valentini
- Division of Radiology, Fondazione IRCCS Policlinico San Matteo, 27100 Pavia, Italy
| | - Paola Franchi
- Department of Diagnostic Radiology, G. Mazzini Hospital, 64100 Teramo, Italy
| | - Giuseppe Cicchetti
- Advanced Radiodiagnostic Center, Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, 00168 Rome, Italy
| | - Gaia Messana
- Diagnostic Imaging Unit, Department of Clinical, Surgical, Diagnostic, and Pediatric Sciences, University of Pavia, 27100 Pavia, Italy
| | - Greta Chiffi
- Secton of Radiology, Department of Radiological and Hematological Sciences, Università Cattolica del Sacro Cuore, 00168 Rome, Italy
| | - Cecilia Strappa
- Secton of Radiology, Department of Radiological and Hematological Sciences, Università Cattolica del Sacro Cuore, 00168 Rome, Italy
| | - Lucio Calandriello
- Advanced Radiodiagnostic Center, Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, 00168 Rome, Italy
| | - Annemilia Del Ciello
- Advanced Radiodiagnostic Center, Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, 00168 Rome, Italy
| | - Alessandra Farchione
- Advanced Radiodiagnostic Center, Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, 00168 Rome, Italy
| | - Lorenzo Preda
- Division of Radiology, Fondazione IRCCS Policlinico San Matteo, 27100 Pavia, Italy
- Diagnostic Imaging Unit, Department of Clinical, Surgical, Diagnostic, and Pediatric Sciences, University of Pavia, 27100 Pavia, Italy
| | - Anna Rita Larici
- Advanced Radiodiagnostic Center, Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, 00168 Rome, Italy
- Secton of Radiology, Department of Radiological and Hematological Sciences, Università Cattolica del Sacro Cuore, 00168 Rome, Italy
| |
Collapse
|
3
|
Katz SI, Straus CM, Roshkovan L, Blyth KG, Frauenfelder T, Gill RR, Lalezari F, Erasmus J, Nowak AK, Gerbaudo VH, Francis RJ, Armato SG. Considerations for Imaging of Malignant Pleural Mesothelioma: A Consensus Statement from the International Mesothelioma Interest Group. J Thorac Oncol 2023; 18:278-298. [PMID: 36549385 DOI: 10.1016/j.jtho.2022.11.018] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/02/2022] [Revised: 11/13/2022] [Accepted: 11/17/2022] [Indexed: 12/24/2022]
Abstract
Malignant pleural mesothelioma (MPM) is an aggressive primary malignancy of the pleura that presents unique radiologic challenges with regard to accurate and reproducible assessment of disease extent at staging and follow-up imaging. By optimizing and harmonizing technical approaches to imaging MPM, the best quality imaging can be achieved for individual patient care, clinical trials, and imaging research. This consensus statement represents agreement on harmonized, standard practices for routine multimodality imaging of MPM, including radiography, computed tomography, 18F-2-deoxy-D-glucose positron emission tomography, and magnetic resonance imaging, by an international panel of experts in the field of pleural imaging assembled by the International Mesothelioma Interest Group. In addition, modality-specific technical considerations and future directions are discussed. A bulleted summary of all technical recommendations is provided.
Collapse
Affiliation(s)
- Sharyn I Katz
- Department of Radiology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania.
| | - Christopher M Straus
- Department of Radiology, University of Chicago Pritzker School of Medicine, Chicago, Illinois
| | - Leonid Roshkovan
- Department of Radiology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania
| | - Kevin G Blyth
- Institute of Cancer Sciences, University of Glasgow, Glasgow, United Kingdom
| | - Thomas Frauenfelder
- Institute for Diagnostic and Interventional Radiology, University Hospital Zurich, Zurich, Switzerland
| | - Ritu R Gill
- Department of Radiology, Beth Israel Lahey Health, Harvard Medical School, Boston, Massachusetts
| | - Ferry Lalezari
- Department of Radiology, Netherlands Cancer Institute, Amsterdam, The Netherlands
| | - Jeremy Erasmus
- Department of Radiology, MD Anderson Cancer Center, Houston, Texas
| | - Anna K Nowak
- Medical School, University of Western Australia, Perth, Australia
| | - Victor H Gerbaudo
- Department of Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts
| | - Roslyn J Francis
- Medical School, University of Western Australia, Perth, Australia; Department of Nuclear Medicine, Sir Charles Gairdner Hospital, Perth, Australia
| | - Samuel G Armato
- Department of Radiology, University of Chicago Pritzker School of Medicine, Chicago, Illinois
| |
Collapse
|
4
|
Chang S, Jung JI, Beck KS. Low Tube Voltage Chest Computed Tomography With Enhancement Using Low-Concentration Iodinated Contrast Media: Comparison of 240 mg/mL Versus 300 mg/mL Iodinated Contrast Media. Can Assoc Radiol J 2023; 74:127-136. [PMID: 35593132 DOI: 10.1177/08465371221102631] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/11/2023] Open
Abstract
Purpose: To evaluate the image quality of low voltage chest computed tomography with enhancement (CECT) using low-concentration-iodine contrast media (LCCM). Method: From 9 December to 19 December 2019, three different protocols were used for 263 patients undergoing chest CECT. Chest CECT was done using routine (300 mgI/ml contrast media with 100 kVp) protocol (group 1), LCCM (240 mgI/ml contrast media)-100 kVp protocol (group 2) and LCCM-80 kVp protocol (group 3) in 91, 97 and 75 patients, respectively. The overall diagnostic acceptability, anatomical depiction, noise and contrast-related artifacts were assessed. Additionally, the mean attenuation, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and figure of merit (FOM) in the aorta and the main pulmonary trunk were measured. Results: The overall diagnostic acceptability scores were not significantly different between groups 1 and 2 (P = .261); group 3 demonstrated significantly lower overall diagnostic acceptability score compared with group 1 (P = .011) or group 2 (P < .001). However, in CECT with iterative reconstruction (IR), the overall diagnostic acceptability scores did not show significant difference among 3 groups. Group 3 showed significantly lower effective radiation dose compared with group 1 (2.33 vs 1.22 mSv, P < .001) or group 2 (2.28 vs .22 mSv, P < .001). Conclusions: In 100 kVp chest CECT, the image quality of using 240 mg/mL iodinated contrast media is comparable to that using 300 mg/mL iodine contrast media, regardless of application of IR; with IR, chest CECT using 80 kVp and 240 mg/mL iodinated contrast media results in acceptable image quality and lower radiation dose.
Collapse
Affiliation(s)
- Suyon Chang
- Department of Radiology, Seoul St Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea
| | - Jung Im Jung
- Department of Radiology, Seoul St Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea
| | - Kyongmin S Beck
- Department of Radiology, Seoul St Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea
| |
Collapse
|
5
|
Henning MK, Aaløkken TM, Johansen S. Contrast medium protocols in routine chest CT: a survey study. Acta Radiol 2022; 63:351-359. [PMID: 33648351 DOI: 10.1177/0284185121997111] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
BACKGROUND Administration of contrast medium (CM) is an important image quality factor in computed tomography (CT) of the chest. There is no clear evidence or guidelines on CM strategies for chest CT, thus a consensus approach is needed. PURPOSE To survey the potential impact on differences in chest CT protocols, with emphasis on strategies for the administration of CM. MATERIAL AND METHODS A total of 170 respondents were included in this survey, which used two different approaches: (i) an online survey was sent to the members of the European Society of Thoracic Imaging (ESTI); and (ii) an email requesting a copy of their CT protocol was sent to all hospitals in Norway, and university hospitals in Sweden and Denmark. The survey focused on factors affecting CM protocols and enhancement in chest CT. RESULTS The overall response rate was 24% (n = 170): 76% of the respondents used a CM concentration of ≥350 mgI/mL; 52% of the respondents used a fixed CM volume strategy. Fixed strategies for injection rate and delay were also the most common approach, practiced by 73% and 57% of the respondents, respectively. The fixed delay was in the range of 20-90 s. Of the respondents, 56% used flexible tube potential strategies (kV). CONCLUSION The chest CT protocols and CM administration strategies employed by the respondents vary widely, affecting the image quality. The results of this study underline the need for further research and consensus guidelines related to chest CT.
Collapse
Affiliation(s)
- Mette Karen Henning
- Faculty of Health Sciences, Department of Life Sciences and Health, Oslo Metropolitan University, Oslo, Norway
- Department of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
| | - Trond Mogens Aaløkken
- Department of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
- Faulty of Medicine, University of Oslo, Oslo, Norway
| | - Safora Johansen
- Faculty of Health Sciences, Department of Life Sciences and Health, Oslo Metropolitan University, Oslo, Norway
- Department of Cancer Treatment, Oslo University Hospital, Oslo, Norway
| |
Collapse
|
6
|
Larici AR, Franchi P, Del Ciello A, Sica G, Coviello D, De Waure C, Cicchetti G, Rovere G, Storto ML, Farchione A, Calandriello L, D'Ambra G, Merlino B, Iezzi R, Marano R, Manfredi R. Role of delayed phase contrast-enhanced CT in the intra-thoracic staging of non-small cell lung cancer (NSCLC): What does it add? Eur J Radiol 2021; 144:109983. [PMID: 34627107 DOI: 10.1016/j.ejrad.2021.109983] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/06/2020] [Revised: 08/20/2021] [Accepted: 09/26/2021] [Indexed: 11/27/2022]
Abstract
PURPOSE The aim of the study was to investigate differences in non-small cell lung cancer (NSCLC) intra-thoracic staging by using contrast-enhanced computed tomography (ce-CT) at the arterial phase (AP), at the arterial plus delayed phases (AP + DEP), and at the delayed phase (DEP), and to evaluate their potential impact on disease staging. MATERIALS AND METHODS Two chest radiologists with different level of expertise and a general radiologist independently reviewed the chest CT exams of 150 patients with NSCLC; CT scans were performed 40 s (AP) and 60 s (DEP) after contrast material injection. Image assessment included three reading sessions: session A (AP), session B (AP + DEP) and session C (DEP). CT descriptors for the primary tumour (T), regional nodal involvement (N), and intra-thoracic metastases (M) were evaluated in each reading session. Readers had to assign a confidence level (CL) for the assessment of each descriptor and define the TNM stage. Friedman and Cochran Q test was used to compare the assessments of the 3 reading sessions; inter-reader agreement was determined (Intraclass Correlation Coefficient - ICC). RESULTS The CL was significantly higher in sessions B and C than in session A for all descriptors, with the exception of pulmonary arterial invasion. Primary tumour inner necrosis and regional nodal involvement were detected in a significantly higher number of cases in sessions B and C as compared to session A (p ≤ 0.001). DEP significantly changed N stage determination (p < 0.001), particularly N3, and excluded chest wall invasion (p = 0.05) and venous invasion (p = 0.001). The agreement was good among the 3 readers (ICC = 0.761) and excellent between the 2 chest radiologists (ICC ≥ 0.940), regardless of the contrast phase. CONCLUSIONS The 60-second DEP ce-CT for staging NSCLC significantly increased the readers' CL, changed the N stage determination, and helped excluding chest wall invasion and venous invasion.
Collapse
Affiliation(s)
- Anna Rita Larici
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy.
| | - Paola Franchi
- Department of Diagnostic Radiology, G. Mazzini Hospital, Teramo, Italy
| | - Annemilia Del Ciello
- Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Giuliano Sica
- Department of Radiology, Azienda ULSS N.1 Dolomiti Presidio Ospedaliero, Feltre e Lamon, Italy
| | - Davide Coviello
- Radiology, Ospedale Valdelsa-Campostaggia, Azienda USL Toscana Sud-Est, Italy
| | - Chiara De Waure
- Department of Experimental Medicine, University of Perugia, Italy
| | - Giuseppe Cicchetti
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Giuseppe Rovere
- Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Maria Luigia Storto
- Bracco Diagnostics Inc, Global Medical and Regulatory Affairs, Monroe Twp, NJ, USA
| | - Alessandra Farchione
- Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Lucio Calandriello
- Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Giulia D'Ambra
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy
| | - Biagio Merlino
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Roberto Iezzi
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Riccardo Marano
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| | - Riccardo Manfredi
- Department of Radiological and Hematological Sciences, Section of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario "A. Gemelli" IRCCS, Rome, Italy
| |
Collapse
|
7
|
Brooke JP, Hall IP. Novel Thoracic MRI Approaches for the Assessment of Pulmonary Physiology and Inflammation. ADVANCES IN EXPERIMENTAL MEDICINE AND BIOLOGY 2021; 1304:123-145. [PMID: 34019267 DOI: 10.1007/978-3-030-68748-9_8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 02/04/2023]
Abstract
Excessive pulmonary inflammation can lead to damage of lung tissue, airway remodelling and established structural lung disease. Novel therapeutics that specifically target inflammatory pathways are becoming increasingly common in clinical practice, but there is yet to be a similar stepwise change in pulmonary diagnostic tools. A variety of thoracic magnetic resonance imaging (MRI) tools are currently in development, which may soon fulfil this emerging clinical need for highly sensitive assessments of lung structure and function. Given conventional MRI techniques are poorly suited to lung imaging, alternate strategies have been developed, including the use of inhaled contrast agents, intravenous contrast and specialized lung MR sequences. In this chapter, we discuss technical challenges of performing MRI of the lungs and how they may be overcome. Key thoracic MRI modalities are reviewed, namely, hyperpolarized noble gas MRI, oxygen-enhanced MRI (OE-MRI), ultrashort echo time (UTE) MRI and dynamic contrast-enhanced (DCE) MRI. Finally, we consider potential clinical applications of these techniques including phenotyping of lung disease, evaluation of novel pulmonary therapeutic efficacy and longitudinal assessment of specific patient groups.
Collapse
Affiliation(s)
- Jonathan P Brooke
- Department of Respiratory Medicine, University of Nottingham, Queens Medical Centre, Nottingham, UK.
| | - Ian P Hall
- Department of Respiratory Medicine, University of Nottingham, Queens Medical Centre, Nottingham, UK.
| |
Collapse
|
8
|
Paluru N, Dayal A, Jenssen HB, Sakinis T, Cenkeramaddi LR, Prakash J, Yalavarthy PK. Anam-Net: Anamorphic Depth Embedding-Based Lightweight CNN for Segmentation of Anomalies in COVID-19 Chest CT Images. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2021; 32:932-946. [PMID: 33544680 PMCID: PMC8544939 DOI: 10.1109/tnnls.2021.3054746] [Citation(s) in RCA: 57] [Impact Index Per Article: 19.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/09/2020] [Revised: 11/14/2020] [Accepted: 01/21/2021] [Indexed: 05/18/2023]
Abstract
Chest computed tomography (CT) imaging has become indispensable for staging and managing coronavirus disease 2019 (COVID-19), and current evaluation of anomalies/abnormalities associated with COVID-19 has been performed majorly by the visual score. The development of automated methods for quantifying COVID-19 abnormalities in these CT images is invaluable to clinicians. The hallmark of COVID-19 in chest CT images is the presence of ground-glass opacities in the lung region, which are tedious to segment manually. We propose anamorphic depth embedding-based lightweight CNN, called Anam-Net, to segment anomalies in COVID-19 chest CT images. The proposed Anam-Net has 7.8 times fewer parameters compared to the state-of-the-art UNet (or its variants), making it lightweight capable of providing inferences in mobile or resource constraint (point-of-care) platforms. The results from chest CT images (test cases) across different experiments showed that the proposed method could provide good Dice similarity scores for abnormal and normal regions in the lung. We have benchmarked Anam-Net with other state-of-the-art architectures, such as ENet, LEDNet, UNet++, SegNet, Attention UNet, and DeepLabV3+. The proposed Anam-Net was also deployed on embedded systems, such as Raspberry Pi 4, NVIDIA Jetson Xavier, and mobile-based Android application (CovSeg) embedded with Anam-Net to demonstrate its suitability for point-of-care platforms. The generated codes, models, and the mobile application are available for enthusiastic users at https://github.com/NaveenPaluru/Segmentation-COVID-19.
Collapse
Affiliation(s)
- Naveen Paluru
- Department of Computational and Data SciencesIndian Institute of ScienceBengaluru560 012India
| | - Aveen Dayal
- Department of Information and Communication TechnologyUniversity of Agder4879GrimstadNorway
| | - Håvard Bjørke Jenssen
- Department of Radiology and Nuclear MedicineOslo University Hospital0372OsloNorway
- Artificial Intelligence AS0553OsloNorway
| | - Tomas Sakinis
- Department of Radiology and Nuclear MedicineOslo University Hospital0372OsloNorway
- Artificial Intelligence AS0553OsloNorway
| | | | - Jaya Prakash
- Department of Instrumentation and Applied PhysicsIndian Institute of ScienceBengaluru560 012India
| | | |
Collapse
|
9
|
Standardizing percutaneous Microwave Ablation in the treatment of Lung Tumors: a prospective multicenter trial (MALT study). Eur Radiol 2020; 31:2173-2182. [PMID: 32997180 DOI: 10.1007/s00330-020-07299-2] [Citation(s) in RCA: 9] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/25/2020] [Revised: 07/27/2020] [Accepted: 09/15/2020] [Indexed: 02/07/2023]
Abstract
OBJECTIVES To prospectively assess reproducibility, safety, and efficacy of microwave ablation (MWA) in the treatment of unresectable primary and secondary pulmonary tumors. METHODS Patients with unresectable primary and metastatic lung tumors up to 4 cm were enrolled in a multicenter prospective clinical trial and underwent CT-guided MWA. Treatments were delivered using pre-defined MW power and duration settings, based on target tumor size and histology classifications. Patients were followed for up to 24 months. Treatment safety, efficacy, and reproducibility were assessed. Ablation volumes were measured at CT scan and compared with ablation volumes obtained on ex vivo bovine liver using equal treatment settings. RESULTS From September 2015 to September 2017, 69 MWAs were performed in 54 patients, achieving technical success in all cases and treatment completion without deviations from the standardized protocol in 61 procedures (88.4%). Immediate post-MWA CT scans showed ablation dimensions smaller by about 25% than in the ex vivo model; however, a remarkable volumetric increase (40%) of the treated area was observed at 1 month post-ablation. No treatment-related deaths nor complications were recorded. Treatments of equal power and duration yielded fairly reproducible ablation dimensions at 48-h post-MWA scans. In comparison with the ex vivo liver model, in vivo ablation sizes were systematically smaller, by about 25%. Overall LPR was 24.7%, with an average TLP of 8.1 months. OS rates at 12 and 24 months were 98.0% and 71.3%, respectively. CONCLUSIONS Percutaneous CT-guided MWA is a reproducible, safe, and effective treatment for malignant lung tumors up to 4 cm in size. KEY POINTS • Percutaneous MWA treatment of primary and secondary lung tumors is a repeatable, safe, and effective therapeutic option. • It provides a fairly reproducible performance on both the long and short axis of the ablation zone. • When using pre-defined treatment duration and power settings according to tumor histology and size, LPR does not increase with increasing tumor size (up to 4 cm) for both primary and metastatic tumors.
Collapse
|
10
|
Dymbe B, Mæland EV, Styve JR, Rusandu A. Individualization of computed tomography protocols for suspected pulmonary embolism: a national investigation of routines. J Int Med Res 2020; 48:300060520918427. [PMID: 32290743 PMCID: PMC7157970 DOI: 10.1177/0300060520918427] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/14/2022] Open
Abstract
Objective Given the extensive use of computed tomography (CT) in radiation-sensitive patients such as pregnant and pediatric patients, and considering the importance of tailoring CT protocols to patient characteristics for both the radiation dose and image quality, this study was performed to investigate the extent to which individualization of CT protocols is practiced across Norway. Methods This cross-sectional study involved collection of CT protocols and administration of a mini-questionnaire to obtain additional information about how CT examinations are individualized. All public hospitals performing CT to detect pulmonary embolism were invited, and 41% participated. Results Tailoring a standard protocol to different patient groups was more common than using dedicated protocols. Most of the available radiation dose-reduction approaches were used. However, implementation of these strategies was not systematic. Children and pregnant patients were examined without using dedicated CT protocols or by using protocol adjustments focusing on radiation dose reduction in 30% and 39% of the hospitals, respectively. Conclusion Practice optimization is needed, especially the development of dedicated CT protocols or guidelines that tailor the existing protocol to pediatric and pregnant patients. Practice might benefit from a more systematic approach to individualization of CT examinations, such as inserting tailoring instructions into CT protocols.
Collapse
Affiliation(s)
- Berit Dymbe
- Department of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
| | - Elisabeth Vespestad Mæland
- Department of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
| | - Jorunn Rønhovde Styve
- Department of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
| | - Albertina Rusandu
- Department of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
| |
Collapse
|
11
|
Wang X, Teng P, Ontiveros A, Goldin JG, Brown MS. High throughput image labeling on chest computed tomography by deep learning. J Med Imaging (Bellingham) 2020; 7:024501. [PMID: 32219151 DOI: 10.1117/1.jmi.7.2.024501] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/30/2019] [Accepted: 02/26/2020] [Indexed: 11/14/2022] Open
Abstract
When mining image data from PACs or clinical trials or processing large volumes of data without curation, the relevant scans must be identified among irrelevant or redundant data. Only images acquired with appropriate technical factors, patient positioning, and physiological conditions may be applicable to a particular image processing or machine learning task. Automatic labeling is important to make big data mining practical by replacing conventional manual review of every single-image series. Digital imaging and communications in medicine headers usually do not provide all the necessary labels and are sometimes incorrect. We propose an image-based high throughput labeling pipeline using deep learning, aimed at identifying scan direction, scan posture, lung coverage, contrast usage, and breath-hold types. They were posed as different classification problems and some of them involved further segmentation and identification of anatomic landmarks. Images of different view planes were used depending on the specific classification problem. All of our models achieved accuracy > 99 % on test set across different tasks using a research database from multicenter clinical trials.
Collapse
Affiliation(s)
- Xiaoyong Wang
- University of California, Los Angeles, Center for Computer Vision and Imaging Biomarkers, Los Angeles, California, United States.,University of California, Los Angeles, Department of Radiological Sciences, Los Angeles, California, United States
| | - Pangyu Teng
- University of California, Los Angeles, Center for Computer Vision and Imaging Biomarkers, Los Angeles, California, United States.,University of California, Los Angeles, Department of Radiological Sciences, Los Angeles, California, United States
| | - Ashley Ontiveros
- University of California, Los Angeles, Center for Computer Vision and Imaging Biomarkers, Los Angeles, California, United States.,University of California, Los Angeles, Department of Radiological Sciences, Los Angeles, California, United States
| | - Jonathan G Goldin
- University of California, Los Angeles, Center for Computer Vision and Imaging Biomarkers, Los Angeles, California, United States.,University of California, Los Angeles, Department of Radiological Sciences, Los Angeles, California, United States
| | - Matthew S Brown
- University of California, Los Angeles, Center for Computer Vision and Imaging Biomarkers, Los Angeles, California, United States.,University of California, Los Angeles, Department of Radiological Sciences, Los Angeles, California, United States
| |
Collapse
|
12
|
Bhalla AS, Das A, Naranje P, Irodi A, Raj V, Goyal A. Imaging protocols for CT chest: A recommendation. Indian J Radiol Imaging 2019; 29:236-246. [PMID: 31741590 PMCID: PMC6857267 DOI: 10.4103/ijri.ijri_34_19] [Citation(s) in RCA: 27] [Impact Index Per Article: 5.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/15/2019] [Revised: 03/06/2019] [Accepted: 03/21/2019] [Indexed: 12/20/2022] Open
Abstract
Computed Tomography (CT) is the mainstay of diagnostic imaging evaluation of thoracic disorders. However, there are a number of CT protocols ranging from a simple non-contrast CT at one end of the spectrum, and CT perfusion as a complex protocol available only on high-end scanners. With the growing diversity, there is a pressing need for radiologists, and clinicians to have a basic understanding of the recommended CT examinations for individual indications. This brief review aims to summarise the currently prevalent CT examination protocols, including their recommended indications, as well as technical specifications for performing them.
Collapse
Affiliation(s)
- Ashu Seith Bhalla
- Department of Radiodiagnosis, All India Institute of Medical Sciences, Vellore, Tamil Nadu, India
| | - Abanti Das
- Department of Radiodiagnosis, Safdarjung Hospital and Vardhaman Mahavir Medical College, Vellore, Tamil Nadu, India
| | - Priyanka Naranje
- Department of Radiodiagnosis, All India Institute of Medical Sciences, Vellore, Tamil Nadu, India
| | - Aparna Irodi
- Department of Radiology, Christian Medical College and Hospital, Vellore, Tamil Nadu, India
| | - Vimal Raj
- Department of Radiology, Narayana Institute of Cardiac Sciences, 258A, Hosur Rd, Bommasandra Industrial Area, Bengaluru, Karnataka, India
| | - Ankur Goyal
- Department of Radiodiagnosis, All India Institute of Medical Sciences, Vellore, Tamil Nadu, India
| |
Collapse
|
13
|
Management of incidental pulmonary nodule in CT: a survey by the Italian College of Chest Radiology. Radiol Med 2019; 124:602-612. [DOI: 10.1007/s11547-019-01011-1] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/03/2018] [Accepted: 02/21/2019] [Indexed: 12/19/2022]
|