1
|
Dehdab R, Brendlin AS, Grözinger G, Almansour H, Brendel JM, Gassenmaier S, Ghibes P, Werner S, Nikolaou K, Afat S. Enhancing Cone-Beam CT Image Quality in TIPSS Procedures Using AI Denoising. Diagnostics (Basel) 2024; 14:1989. [PMID: 39272773 PMCID: PMC11394631 DOI: 10.3390/diagnostics14171989] [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: 08/14/2024] [Revised: 08/30/2024] [Accepted: 09/02/2024] [Indexed: 09/15/2024] Open
Abstract
Purpose: This study evaluates a deep learning-based denoising algorithm to improve the trade-off between radiation dose, image noise, and motion artifacts in TIPSS procedures, aiming for shorter acquisition times and reduced radiation with maintained diagnostic quality. Methods: In this retrospective study, TIPSS patients were divided based on CBCT acquisition times of 6 s and 3 s. Traditional weighted filtered back projection (Original) and an AI denoising algorithm (AID) were used for image reconstructions. Objective assessments of image quality included contrast, noise levels, and contrast-to-noise ratios (CNRs) through place-consistent region-of-interest (ROI) measurements across various critical areas pertinent to the TIPSS procedure. Subjective assessments were conducted by two blinded radiologists who evaluated the overall image quality, sharpness, contrast, and motion artifacts for each dataset combination. Statistical significance was determined using a mixed-effects model (p ≤ 0.05). Results: From an initial cohort of 60 TIPSS patients, 44 were selected and paired. The mean dose-area product (DAP) for the 6 s acquisitions was 5138.50 ± 1325.57 µGy·m2, significantly higher than the 2514.06 ± 691.59 µGym2 obtained for the 3 s series. CNR was highest in the 6 s-AID series (p < 0.05). Both denoised and original series showed consistent contrast for 6 s and 3 s acquisitions, with no significant noise differences between the 6 s Original and 3 s AID images (p > 0.9). Subjective assessments indicated superior quality in 6 s-AID images, with no significant overall quality difference between the 6 s-Original and 3 s-AID series (p > 0.9). Conclusions: The AI denoising algorithm enhances CBCT image quality in TIPSS procedures, allowing for shorter scans that reduce radiation exposure and minimize motion artifacts.
Collapse
Affiliation(s)
- Reza Dehdab
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Andreas S Brendlin
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Gerd Grözinger
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Haidara Almansour
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Jan Michael Brendel
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Sebastian Gassenmaier
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Patrick Ghibes
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Sebastian Werner
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Konstantin Nikolaou
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| | - Saif Afat
- Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, D-72076 Tuebingen, Germany
| |
Collapse
|
2
|
Brendlin AS, Dehdab R, Stenzl B, Mueck J, Ghibes P, Groezinger G, Kim J, Afat S, Artzner C. Novel Deep Learning Denoising Enhances Image Quality and Lowers Radiation Exposure in Interventional Bronchial Artery Embolization Cone Beam CT. Acad Radiol 2024; 31:2144-2155. [PMID: 37989681 DOI: 10.1016/j.acra.2023.11.003] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/20/2023] [Revised: 10/23/2023] [Accepted: 11/01/2023] [Indexed: 11/23/2023]
Abstract
OBJECTIVES In interventional bronchial artery embolization (BAE), periprocedural cone beam CT (CBCT) improves guiding and localization. However, a trade-off exists between 6-second runs (high radiation dose and motion artifacts, but low noise) and 3-second runs (vice versa). This study aimed to determine the efficacy of an advanced deep learning denoising (DLD) technique in mitigating the trade-offs related to radiation dose and image quality during interventional BAE CBCT. MATERIALS AND METHODS This study included BMI-matched patients undergoing 6-second and 3-second BAE CBCT scans. The dose-area product values (DAP) were obtained. All datasets were reconstructed using standard weighted filtered back projection (OR) and a novel DLD software. Objective image metrics were derived from place-consistent regions of interest, including CT numbers of the Aorta and lung, noise, and contrast-to-noise ratio. Three blinded radiologists performed subjective assessments regarding image quality, sharpness, contrast, and motion artifacts on all dataset combinations in a forced-choice setup (-1 = inferior, 0 = equal; 1 = superior). The points were averaged per item for a total score. Statistical analysis ensued using a properly corrected mixed-effects model with post hoc pairwise comparisons. RESULTS Sixty patients were assessed in 30 matched pairs (age 64 ± 15 years; 10 female). The mean DAP for the 6 s and 3 s runs was 2199 ± 185 µGym² and 1227 ± 90 µGym², respectively. Neither low-dose imaging nor the reconstruction method introduced a significant HU shift (p ≥ 0.127). The 3 s-DLD presented the least noise and superior contrast-to-noise ratio (CNR) (p < 0.001). While subjective evaluation revealed no noticeable distinction between 6 s-DLD and 3 s-DLD in terms of quality (p ≥ 0.996), both outperformed the OR variants (p < 0.001). The 3 s datasets exhibited fewer motion artifacts than the 6 s datasets (p < 0.001). CONCLUSIONS DLD effectively mitigates the trade-off between radiation dose, image noise, and motion artifact burden in regular reconstructed BAE CBCT by enabling diagnostic scans with low radiation exposure and inherently low motion artifact burden at short examination times.
Collapse
Affiliation(s)
- Andreas S Brendlin
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.).
| | - Reza Dehdab
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| | - Benedikt Stenzl
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| | - Jonas Mueck
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| | - Patrick Ghibes
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| | - Gerd Groezinger
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| | - Jonghyo Kim
- Department of Radiology, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea (J.K.); ClariPi Inc., 11 Ihwajang 1-gil, Jongno-gu, Seoul 03088, Republic of Korea (J.K.)
| | - Saif Afat
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| | - Christoph Artzner
- Department of Diagnostic and Interventional Radiology, Eberhard-Karls University, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany (A.S.B., R.D., B.S., J.M., P.G., G.G., S.A., C.A.)
| |
Collapse
|
3
|
Sadia RT, Chen J, Zhang J. CT image denoising methods for image quality improvement and radiation dose reduction. J Appl Clin Med Phys 2024; 25:e14270. [PMID: 38240466 PMCID: PMC10860577 DOI: 10.1002/acm2.14270] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/18/2023] [Revised: 12/15/2023] [Accepted: 12/28/2023] [Indexed: 02/13/2024] Open
Abstract
With the ever-increasing use of computed tomography (CT), concerns about its radiation dose have become a significant public issue. To address the need for radiation dose reduction, CT denoising methods have been widely investigated and applied in low-dose CT images. Numerous noise reduction algorithms have emerged, such as iterative reconstruction and most recently, deep learning (DL)-based approaches. Given the rapid advancements in Artificial Intelligence techniques, we recognize the need for a comprehensive review that emphasizes the most recently developed methods. Hence, we have performed a thorough analysis of existing literature to provide such a review. Beyond directly comparing the performance, we focus on pivotal aspects, including model training, validation, testing, generalizability, vulnerability, and evaluation methods. This review is expected to raise awareness of the various facets involved in CT image denoising and the specific challenges in developing DL-based models.
Collapse
Affiliation(s)
- Rabeya Tus Sadia
- Department of Computer ScienceUniversity of KentuckyLexingtonKentuckyUSA
| | - Jin Chen
- Department of Medicine‐NephrologyUniversity of Alabama at BirminghamBirminghamAlabamaUSA
| | - Jie Zhang
- Department of RadiologyUniversity of KentuckyLexingtonKentuckyUSA
| |
Collapse
|
4
|
Kawauchi S, Chida K, Hamada Y, Tsuruta W. Image Quality and Radiation Dose of Conventional and Wide-Field High-Resolution Cone-Beam Computed Tomography for Cerebral Angiography: A Phantom Study. Tomography 2023; 9:1683-1693. [PMID: 37736987 PMCID: PMC10514806 DOI: 10.3390/tomography9050134] [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] [Received: 06/20/2023] [Revised: 08/24/2023] [Accepted: 08/29/2023] [Indexed: 09/23/2023] Open
Abstract
There has been an increase in the use of interventional neuroradiology procedures because of their non-invasiveness compared to surgeries and the improved image quality of fluoroscopy, digital subtraction angiography, and rotational angiography. Although cone-beam computed tomography (CBCT) images are inferior to multi-detector CT images in terms of low-contrast detectability and lower radiation doses, CBCT scans are frequently performed because of their accessibility. This study aimed to evaluate the image quality and radiation dose of two different high-resolution CBCTs (HR CBCT): conventional (C-HR CBCT) and wide-field HR CBCT (W-HR CBCT). The modulation transfer function (MTF), noise power spectrum (NPS), and contrast-to-noise ratio (CNR) were used to evaluate the image quality. On comparing the MTF of C-HR CBCT with a 256 × 256 matrix and that of W-HR CBCT with a 384 × 384 matrix, the MTF of W-HR CBCT with the 384 × 384 matrix was larger. A comparison of the NPS and CNR of C-HR CBCT with a 256 × 256 matrix and W-HR CBCT with a 384 × 384 matrix showed that both values were comparable. The reference air kerma values were equal for C-HR CBCT and W-HR CBCT; however, the value of the kerma area product was 1.44 times higher for W-HR CBCT compared to C-HR CBCT. The W-HR CBCT allowed for improved spatial resolution while maintaining the image noise and low-contrast detectability by changing the number of image matrices from 256 × 256 to 384 × 384. Our study revealed the image characteristics and radiation dose of W-HR CBCT. Given its advantages of low-contrast detectability and wide-area imaging with high spatial resolution, W-HR CBCT may be useful in interventional neuroradiology for acute ischemic stroke.
Collapse
Affiliation(s)
- Satoru Kawauchi
- Department of Radiology, Toranomon Hospital, 2-2-2 Toranomon, Minato-ku, Tokyo 105-8470, Japan; (S.K.); (Y.H.)
- Department of Radiological Technology, Tohoku University Graduate School of Medicine, 2-1 Seiryo, Aoba-ku, Sendai 980-8575, Miyagi, Japan
- Okinaka Memorial Institute for Medical Research, 2-2-2 Toranomon, Minato-ku, Tokyo 105-8470, Japan
| | - Koichi Chida
- Department of Radiological Technology, Tohoku University Graduate School of Medicine, 2-1 Seiryo, Aoba-ku, Sendai 980-8575, Miyagi, Japan
- Department of Radiation Disaster Medicine, International Research Institute of Disaster Science, Tohoku University, 468-1 Aramaki Aza-Aoba, Aoba-ku, Sendai 980-0845, Miyagi, Japan
| | - Yusuke Hamada
- Department of Radiology, Toranomon Hospital, 2-2-2 Toranomon, Minato-ku, Tokyo 105-8470, Japan; (S.K.); (Y.H.)
| | - Wataro Tsuruta
- Department of Endovascular Neurosurgery, Toranomon Hospital, 2-2-2 Toranomon, Minato-ku, Tokyo 105-8470, Japan;
| |
Collapse
|
5
|
Terzis R, Reimer RP, Nelles C, Celik E, Caldeira L, Heidenreich A, Storz E, Maintz D, Zopfs D, Große Hokamp N. Deep-Learning-Based Image Denoising in Imaging of Urolithiasis: Assessment of Image Quality and Comparison to State-of-the-Art Iterative Reconstructions. Diagnostics (Basel) 2023; 13:2821. [PMID: 37685359 PMCID: PMC10486912 DOI: 10.3390/diagnostics13172821] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/07/2023] [Revised: 08/23/2023] [Accepted: 08/25/2023] [Indexed: 09/10/2023] Open
Abstract
This study aimed to compare the image quality and diagnostic accuracy of deep-learning-based image denoising reconstructions (DLIDs) to established iterative reconstructed algorithms in low-dose computed tomography (LDCT) of patients with suspected urolithiasis. LDCTs (CTDIvol, 2 mGy) of 76 patients (age: 40.3 ± 5.2 years, M/W: 51/25) with suspected urolithiasis were retrospectively included. Filtered-back projection (FBP), hybrid iterative and model-based iterative reconstruction (HIR/MBIR, respectively) were reconstructed. FBP images were processed using a Food and Drug Administration (FDA)-approved DLID. ROIs were placed in renal parenchyma, fat, muscle and urinary bladder. Signal- and contrast-to-noise ratios (SNR/CNR, respectively) were calculated. Two radiologists evaluated image quality on five-point Likert scales and urinary stones. The results showed a progressive decrease in image noise from FBP, HIR and DLID to MBIR with significant differences between each method (p < 0.05). SNR and CNR were comparable between MBIR and DLID, while it was significantly lower in HIR followed by FBP (e.g., SNR: 1.5 ± 0.3; 1.4 ± 0.4; 1.0 ± 0.3; 0.7 ± 0.2, p < 0.05). Subjective analysis confirmed best image quality in MBIR, followed by DLID and HIR, both being superior to FBP (p < 0.05). Diagnostic accuracy for urinary stone detection was best using MBIR (0.94), lowest using FBP (0.84) and comparable between DLID (0.90) and HIR (0.90). Stone size measurements were consistent between all reconstructions and showed excellent correlation (r2 = 0.958-0.975). In conclusion, MBIR yielded the highest image quality and diagnostic accuracy, with DLID producing better results than HIR and FBP in image quality and matching HIR in diagnostic precision.
Collapse
Affiliation(s)
- Robert Terzis
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - Robert Peter Reimer
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - Christian Nelles
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - Erkan Celik
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - Liliana Caldeira
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - Axel Heidenreich
- Department of Urology, Uro-Oncology, Robot-Assisted and Specialized Urologic Surger, University Hospital Cologne, 50937 Cologne, Germany
| | - Enno Storz
- Department of Urology, Uro-Oncology, Robot-Assisted and Specialized Urologic Surger, University Hospital Cologne, 50937 Cologne, Germany
| | - David Maintz
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - David Zopfs
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| | - Nils Große Hokamp
- Institute for Diagnostic and Interventional Radiology, University Hospital Cologne, 50937 Cologne, Germany (D.M.); (D.Z.)
| |
Collapse
|