1
|
Panthier F, Melchionna A, Crawford-Smith H, Phillipou Y, Choong S, Arumuham V, Allen S, Allen C, Smith D. Can Artificial Intelligence Accurately Detect Urinary Stones? A Systematic Review. J Endourol 2024; 38:725-740. [PMID: 38666692 DOI: 10.1089/end.2023.0717] [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: 06/01/2024] Open
Abstract
Objectives: To perform a systematic review on artificial intelligence (AI) performances to detect urinary stones. Methods: A PROSPERO-registered (CRD473152) systematic search of Scopus, Web of Science, Embase, and PubMed databases was performed to identify original research articles pertaining to AI stone detection or measurement, using search terms ("automatic" OR "machine learning" OR "convolutional neural network" OR "artificial intelligence" OR "detection" AND "stone volume"). Risk-of-bias (RoB) assessment was performed according to the Cochrane RoB tool, the Joanna Briggs Institute Checklist for nonrandomized studies, and the Checklist for Artificial Intelligence in Medical Imaging (CLAIM). Results: Twelve studies were selected for the final review, including three multicenter and nine single-center retrospective studies. Eleven studies completed at least 50% of the CLAIM checkpoints and only one presented a high RoB. All included studies aimed to detect kidney (5/12, 42%), ureter (2/12, 16%), or urinary (5/12, 42%) stones on noncontrast computed tomography (NCCT), but 42% intended to automate measurement. Stone distinction from vascular calcification interested two studies. All studies used AI machine learning network training and internal validation, but a single one provided an external validation. Trained networks achieved stone detection, with sensitivity, specificity, and accuracy rates ranging from 58.7% to 100%, 68.5% to 100%, and 63% to 99.95%, respectively. Detection Dice score ranged from 83% to 97%. A high correlation between manual and automated stone volume (r = 0.95) was noted. Differentiate distal ureteral stones and phleboliths seemed feasible. Conclusions: AI processes can achieve automated urinary stone detection from NCCT. Further studies should provide urinary stone detection coupled with phlebolith distinction and an external validation, and include anatomical abnormalities and urologic foreign bodies (ureteral stent and nephrostomy tubes) cases.
Collapse
Affiliation(s)
- Frédéric Panthier
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
- Sorbonne University GRC Urolithiasis No. 20 Tenon Hospital, Paris, France
- Progressive Endourological Association for Research and Leading Solutions (PEARLS), Paris, France
- PIMM, UMR 8006 CNRS-Arts et Métiers ParisTech, Paris, France
| | - Alberto Melchionna
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
| | - Hugh Crawford-Smith
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
| | - Yiannis Phillipou
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
| | - Simon Choong
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
| | - Vimoshan Arumuham
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
| | - Sian Allen
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
| | - Clare Allen
- Department of Radiology, University College London Hospitals, London, United Kingdom
| | - Daron Smith
- Department of Urology, Westmoreland Street Hospital, UCLH NHS Foundation Trust, London, United Kingdom
- Endourology Academy
- Social Media Committee, Endourological Society
| |
Collapse
|
2
|
Aortic CT angiography dose reduction: investigation of optimal noise index and iterative algorithm strength in combination with low kV. Radiol Med 2015; 121:291-300. [DOI: 10.1007/s11547-015-0611-4] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/12/2015] [Accepted: 11/26/2015] [Indexed: 12/12/2022]
|
3
|
Kubo T, Ohno Y, Kauczor HU, Hatabu H. Radiation dose reduction in chest CT--review of available options. Eur J Radiol 2014; 83:1953-61. [PMID: 25066756 DOI: 10.1016/j.ejrad.2014.06.033] [Citation(s) in RCA: 51] [Impact Index Per Article: 5.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/29/2013] [Revised: 06/24/2014] [Accepted: 06/30/2014] [Indexed: 10/25/2022]
Abstract
Computed tomography currently accounts for the majority of radiation exposure related to medical imaging. Although technological improvement of CT scanners has reduced the radiation dose of individual examinations, the benefit was overshadowed by the rapid increase in the number of CT examinations. Radiation exposure from CT examination should be kept as low as reasonably possible for patient safety. Measures to avoid inappropriate CT examinations are needed. Principles and information on radiation dose reduction in chest CT are reviewed in this article. The reduction of tube current and tube potential are the mainstays of dose reduction methods. Study results indicate that routine protocols with reduced tube current are feasible with diagnostic results comparable to conventional standard dose protocols. Tube current adjustment is facilitated by the advent of automatic tube current modulation systems by setting the appropriate image quality level for the purpose of the examination. Tube potential reduction is an effective method for CT pulmonary angiography. Tube potential reduction often requires higher tube current for satisfactory image quality, but may still contribute to significant radiation dose reduction. Use of lower tube potential also has considerable advantage for smaller patients. Improvement in image production, especially the introduction of iterative reconstruction methods, is expected to lower radiation dose significantly. Radiation dose reduction in CT is a multifaceted issue. Understanding these aspects leads to an optimal solution for various indications of chest CT.
Collapse
Affiliation(s)
- Takeshi Kubo
- Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan.
| | - Yoshiharu Ohno
- Department of Radiology, Kobe University Graduate School of Medicine, 7-5-2, Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan.
| | - Hans Ulrich Kauczor
- Diagnostic and Interventional Radiology, University Clinic Heidelberg, Im Neuenheimer Feld 110, D-69120 Heidelberg, Germany.
| | - Hiroto Hatabu
- Department of Radiology, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, United States.
| |
Collapse
|
4
|
Greffier J, Fernandez A, Macri F, Freitag C, Metge L, Beregi JP. Which dose for what image? Iterative reconstruction for CT scan. Diagn Interv Imaging 2013; 94:1117-21. [DOI: 10.1016/j.diii.2013.03.008] [Citation(s) in RCA: 39] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/01/2022]
|
5
|
Tsapaki V, Rehani M, Saini S. Radiation safety in abdominal computed tomography. Semin Ultrasound CT MR 2010; 31:29-38. [PMID: 20102693 DOI: 10.1053/j.sult.2009.09.004] [Citation(s) in RCA: 30] [Impact Index Per Article: 2.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/12/2023]
Abstract
The rapid technological developments in computed tomography (CT) have enabled many new clinical applications in the abdominal region. Abdomen CT is considered as a high radiation dose examination due to the large number of radiosensitive organs in the field of view. CT radiation dose has received a lot of attention not only by the medical specialties and researchers, but also by patients and media. This article reviews the situation on radiation dose and risk and provides practical guidelines to effectively manage the radiation dose without losing the benefits and maintaining diagnostic confidence in CT procedures.
Collapse
|
6
|
Abstract
OBJECTIVE This article aims to summarize the available data on reducing radiation dose exposure in routine chest CT protocols. First, the general aspects of radiation dose in CT and radiation risk are discussed, followed by the effect of changing parameters on image quality. Finally, the results of previous radiation dose reduction studies are reviewed, and important information contributing to radiation dose reduction will be shared. CONCLUSION A variety of methods and techniques for radiation dose reduction should be used to ensure that radiation exposure is kept as low as is reasonably achievable.
Collapse
|
7
|
Brisse HJ, Madec L, Gaboriaud G, Lemoine T, Savignoni A, Neuenschwander S, Aubert B, Rosenwald JC. Automatic exposure control in multichannel CT with tube current modulation to achieve a constant level of image noise: Experimental assessment on pediatric phantoms. Med Phys 2007; 34:3018-33. [PMID: 17822010 DOI: 10.1118/1.2746492] [Citation(s) in RCA: 42] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/12/2022] Open
Abstract
Automatic exposure control (AEC) systems have been developed by computed tomography (CT) manufacturers to improve the consistency of image quality among patients and to control the absorbed dose. Since a multichannel helical CT scan may easily increase individual radiation doses, this technical improvement is of special interest in children who are particularly sensitive to ionizing radiation, but little information is currently available regarding the precise performance of these systems on small patients. Our objective was to assess an AEC system on pediatric dose phantoms by studying the impact of phantom transmission and acquisition parameters on tube current modulation, on the resulting absorbed dose and on image quality. We used a four-channel CT scan working with a patient-size and z-axis-based AEC system designed to achieve a constant noise within the reconstructed images by automatically adjusting the tube current during acquisition. The study was performed with six cylindrical poly(methylmethacrylate) (PMMA) phantoms of variable diameters (10-32 cm) and one 5 years of age equivalent pediatric anthropomorphic phantom. After a single scan projection radiograph (SPR), helical acquisitions were performed and images were reconstructed with a standard convolution kernel. Tube current modulation was studied with variable SPR settings (tube angle, mA, kVp) and helical parameters (6-20 HU noise indices, 80-140 kVp tube potential, 0.8-4 s. tube rotation time, 5-20 mm x-ray beam thickness, 0.75-1.5 pitch, 1.25-10 mm image thickness, variable acquisition, and reconstruction fields of view). CT dose indices (CTDIvol) were measured, and the image quality criterion used was the standard deviation of the CT number measured in reconstructed images of PMMA material. Observed tube current levels were compared to the expected values from Brooks and Di Chiro's [R.A. Brooks and G.D. Chiro, Med. Phys. 3, 237-240 (1976)] model and calculated values (product of a reference value multiplied by a dose ratio measured with thermoluminescent dosimeters). Our study demonstrates that this AEC system accurately modulates the tube current according to phantom size and transmission to achieve a stable image noise. The system accurately controls the tube current when changing tube rotation time, tube potential, or image thickness, with minimal variations of the resulting noise. Nevertheless, CT users should be aware of possible changes of tube current and resulting dose and quality according to several parameters: the tube angle and tube potential used for SPR, the x-ray beam thickness (tube current decreases and image noise increases when doubling x-ray beam thickness), the pitch value (a pitch decrease leads to a higher dose but also to a higher noise), and the acquisition field of view (FOV) (tube current is lower when using the small acquisition FOV compared to the large one, but the use of small acquisition FOV at 120 kVp leads to a peculiar increase of tube current and CTDIvol).
Collapse
|
8
|
Cordoliani YS. [Radiation protection]. JOURNAL DE RADIOLOGIE 2005; 86:890-1. [PMID: 16342871 DOI: 10.1016/s0221-0363(05)81464-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/05/2023]
Affiliation(s)
- Y S Cordoliani
- Service de Radiologie, HIA du Val-de-Grâce, 74 boulevard du Port-Royal, 75230 Paris 5.
| |
Collapse
|