1
|
Wells JB, Lewis SJ, Barron M, Trieu PD. Surgical and Radiology Trainees' Proficiency in Reading Mammograms: the Importance of Education for Cancer Localisation. JOURNAL OF CANCER EDUCATION : THE OFFICIAL JOURNAL OF THE AMERICAN ASSOCIATION FOR CANCER EDUCATION 2024; 39:186-193. [PMID: 38100062 PMCID: PMC10994868 DOI: 10.1007/s13187-023-02393-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Accepted: 12/03/2023] [Indexed: 04/05/2024]
Abstract
Medical imaging with mammography plays a very important role in screening and diagnosis of breast cancer, Australia's most common female cancer. The visualisation of cancers on mammograms often forms a diagnosis and guidance for radiologists and breast surgeons, and education platforms that provide real cases in a simulated testing environment have been shown to improve observer performance for radiologists. This study reports on the performance of surgical and radiology trainees in locating breast cancers. An enriched test set of 20 mammography cases (6 cancer and 14 cancer free) was created, and 18 surgical trainees and 32 radiology trainees reviewed the cases via the Breast Screen Reader Assessment Strategy (BREAST) platform and marked any lesions identifiable. Further analysis of performance with high- and low-density cases was undertaken, and standard metrics including sensitivity and specificity. Radiology trainees performed significantly better than surgical trainees in terms of specificity (0.72 vs. 0.35; P < 0.01). No significant differences were observed between the surgical and radiology trainees in sensitivity or lesion sensitivity. Mixed results were obtained with participants regarding breast density, with higher density cases generally having lower performance. The higher specificity of the radiology trainees compared to the surgical trainees likely represents less exposure to negative mammography cases. The use of high-fidelity simulated self-test environments like BREAST is able to benchmark, understand and build strategies for improving cancer education in a safe environment, including identifying challenging scenarios like breast density for enhanced training.
Collapse
Affiliation(s)
- J B Wells
- Discipline of Medical Imaging Sciences, Faculty of Medicine and Health, University of Sydney, D18 Susan Wakil Health Building, Western Avenue, Camperdown, NSW, 2006, Australia
| | - S J Lewis
- Discipline of Medical Imaging Sciences, Faculty of Medicine and Health, University of Sydney, D18 Susan Wakil Health Building, Western Avenue, Camperdown, NSW, 2006, Australia.
| | - M Barron
- Discipline of Medical Imaging Sciences, Faculty of Medicine and Health, University of Sydney, D18 Susan Wakil Health Building, Western Avenue, Camperdown, NSW, 2006, Australia
| | - P D Trieu
- Discipline of Medical Imaging Sciences, Faculty of Medicine and Health, University of Sydney, D18 Susan Wakil Health Building, Western Avenue, Camperdown, NSW, 2006, Australia
| |
Collapse
|
2
|
Siviengphanom S, Gandomkar Z, Lewis SJ, Brennan PC. Global Radiomic Features from Mammography for Predicting Difficult-To-Interpret Normal Cases. J Digit Imaging 2023; 36:1541-1552. [PMID: 37253894 PMCID: PMC10406750 DOI: 10.1007/s10278-023-00836-7] [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: 11/28/2022] [Revised: 04/05/2023] [Accepted: 04/13/2023] [Indexed: 06/01/2023] Open
Abstract
This work aimed to investigate whether global radiomic features (GRFs) from mammograms can predict difficult-to-interpret normal cases (NCs). Assessments from 537 readers interpreting 239 normal mammograms were used to categorise cases as 120 difficult-to-interpret and 119 easy-to-interpret based on cases having the highest and lowest difficulty scores, respectively. Using lattice- and squared-based approaches, 34 handcrafted GRFs per image were extracted and normalised. Three classifiers were constructed: (i) CC and (ii) MLO using the GRFs from corresponding craniocaudal and mediolateral oblique images only, based on the random forest technique for distinguishing difficult- from easy-to-interpret NCs, and (iii) CC + MLO using the median predictive scores from both CC and MLO models. Useful GRFs for the CC and MLO models were recognised using a scree test. The CC and MLO models were trained and validated using the leave-one-out-cross-validation. The models' performances were assessed by the AUC and compared using the DeLong test. A Kruskal-Wallis test was used to examine if the 34 GRFs differed between difficult- and easy-to-interpret NCs and if difficulty level based on the traditional breast density (BD) categories differed among 115 low-BD and 124 high-BD NCs. The CC + MLO model achieved higher performance (0.71 AUC) than the individual CC and MLO model alone (0.66 each), but statistically non-significant difference was found (all p > 0.05). Six GRFs were identified to be valuable in describing difficult-to-interpret NCs. Twenty features, when compared between difficult- and easy-to-interpret NCs, differed significantly (p < 0.05). No statistically significant difference was observed in difficulty between low- and high-BD NCs (p = 0.709). GRF mammographic analysis can predict difficult-to-interpret NCs.
Collapse
Affiliation(s)
- Somphone Siviengphanom
- Medical Image Optimisation and Perception Group, Discipline of Medical Imaging Science, Sydney School of Health Sciences, Faculty of Medicine and Health, the University of Sydney, Sydney, NSW, 2006, Australia.
| | - Ziba Gandomkar
- Medical Image Optimisation and Perception Group, Discipline of Medical Imaging Science, Sydney School of Health Sciences, Faculty of Medicine and Health, the University of Sydney, Sydney, NSW, 2006, Australia
| | - Sarah J Lewis
- Medical Image Optimisation and Perception Group, Discipline of Medical Imaging Science, Sydney School of Health Sciences, Faculty of Medicine and Health, the University of Sydney, Sydney, NSW, 2006, Australia
| | - Patrick C Brennan
- Medical Image Optimisation and Perception Group, Discipline of Medical Imaging Science, Sydney School of Health Sciences, Faculty of Medicine and Health, the University of Sydney, Sydney, NSW, 2006, Australia
| |
Collapse
|
3
|
Tao X, Gandomkar Z, Li T, Brennan PC, Reed W. Using Radiomics-Based Machine Learning to Create Targeted Test Sets to Improve Specific Mammography Reader Cohort Performance: A Feasibility Study. J Pers Med 2023; 13:888. [PMID: 37373877 DOI: 10.3390/jpm13060888] [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: 02/25/2023] [Revised: 05/11/2023] [Accepted: 05/23/2023] [Indexed: 06/29/2023] Open
Abstract
Mammography interpretation is challenging with high error rates. This study aims to reduce the errors in mammography reading by mapping diagnostic errors against global mammographic characteristics using a radiomics-based machine learning approach. A total of 36 radiologists from cohort A (n = 20) and cohort B (n = 16) read 60 high-density mammographic cases. Radiomic features were extracted from three regions of interest (ROIs), and random forest models were trained to predict diagnostic errors for each cohort. Performance was evaluated using sensitivity, specificity, accuracy, and AUC. The impact of ROI placement and normalization on prediction was investigated. Our approach successfully predicted both the false positive and false negative errors of both cohorts but did not consistently predict location errors. The errors produced by radiologists from cohort B were less predictable compared to those in cohort A. The performance of the models did not show significant improvement after feature normalization, despite the mammograms being produced by different vendors. Our novel radiomics-based machine learning pipeline focusing on global radiomic features could predict false positive and false negative errors. The proposed method can be used to develop group-tailored mammographic educational strategies to help improve future mammography reader performance.
Collapse
Affiliation(s)
- Xuetong Tao
- Discipline of Medical Imaging Science, Faculty of Health Sciences, The University of Sydney, Sydney, NSW 2006, Australia
| | - Ziba Gandomkar
- Discipline of Medical Imaging Science, Faculty of Health Sciences, The University of Sydney, Sydney, NSW 2006, Australia
| | - Tong Li
- The Daffodil Centre, The University of Sydney, Sydney, NSW 2006, Australia
- Sydney School of Public Health, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW 2006, Australia
| | - Patrick C Brennan
- Discipline of Medical Imaging Science, Faculty of Health Sciences, The University of Sydney, Sydney, NSW 2006, Australia
| | - Warren Reed
- Discipline of Medical Imaging Science, Faculty of Health Sciences, The University of Sydney, Sydney, NSW 2006, Australia
| |
Collapse
|
4
|
Trieu PD(Y, Borecky N, Li T, Brennan PC, Barron ML, Lewis SJ. The Impact of Prior Mammograms on the Diagnostic Performance of Radiologists in Early Breast Cancer Detection: A Focus on Breast Density, Lesion Features and Vendors Using Wholly Digital Screening Cases. Cancers (Basel) 2023; 15:cancers15041339. [PMID: 36831680 PMCID: PMC9954188 DOI: 10.3390/cancers15041339] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/31/2022] [Revised: 02/09/2023] [Accepted: 02/12/2023] [Indexed: 02/22/2023] Open
Abstract
BACKGROUND This study aims to investigate the diagnostic efficacy of radiologists when reading screening mammograms in the absence of previous images, and with the presence of prior images from the same and different vendors. METHODS 612 radiologists' readings across 9 test sets, consisting of 540 screening mammograms (361-normal and 179-cancer) with 245 cases having prior images obtained from same vendor as current images, 129 from a different vendor and 166 cases having no prior images, were retrospectively analysed. True positive (sensitivity), true negative (specificity) and area under ROC curve (AUC) values of radiologists were calculated for three groups of cases (without prior images (NP), with prior images from same vendor (SP), and with prior images from different vendor (DP)). Logistic regression was used to estimate the odds ratio (OR) of true positive, true negative and true cancer localization among case groups with different levels of breast density and lesion characteristics. RESULTS Radiologists obtained 12.8% and 10.3% higher sensitivity in NP and DP than SP (0.803-and-0.785 vs. 0.712; p < 0.0001). Specificity in NP and DP cases were 4.8% and 2.0% lower than SP cases (0.749 and 0.771 vs. 0.787). The AUC values for NP and DP were significantly higher than SP cases across different levels of breast density (0.814-and-0.820 vs. 0.782; p < 0.0001). The odds ratio (OR) of true positive for NP relative to SP was 1.6 (p < 0.0001) and DP relative to SP was 1.5 (p < 0.0001). Radiologists were more like to detect architectural distortion in DP than SP cases (OR = 3.2; p < 0.0001), whilst the OR for abnormal calcifications was 2.85 (p < 0.0001). CONCLUSIONS Cases without previous mammograms or with prior mammograms obtained from different vendors were more likely to benefit radiologists in cancer detection, whilst prior mammograms undertaken from the same vendor were more useful for radiologists in evaluating normal cases.
Collapse
Affiliation(s)
- Phuong Dung (Yun) Trieu
- Department of Clinical Imaging, Faculty of Medicine and Health, The University of Sydney, Level 7-D18, Susan Wakil Health Building, Camperdown, NSW 2006, Australia; (N.B.); (T.L.); (P.C.B.); (M.L.B.); (S.J.L.)
- Correspondence:
| | - Natacha Borecky
- Department of Clinical Imaging, Faculty of Medicine and Health, The University of Sydney, Level 7-D18, Susan Wakil Health Building, Camperdown, NSW 2006, Australia; (N.B.); (T.L.); (P.C.B.); (M.L.B.); (S.J.L.)
- BreastScreen New South Wales (North Coast), Lismore, NSW P.O. Box 1098, Australia
| | - Tong Li
- Department of Clinical Imaging, Faculty of Medicine and Health, The University of Sydney, Level 7-D18, Susan Wakil Health Building, Camperdown, NSW 2006, Australia; (N.B.); (T.L.); (P.C.B.); (M.L.B.); (S.J.L.)
| | - Patrick C. Brennan
- Department of Clinical Imaging, Faculty of Medicine and Health, The University of Sydney, Level 7-D18, Susan Wakil Health Building, Camperdown, NSW 2006, Australia; (N.B.); (T.L.); (P.C.B.); (M.L.B.); (S.J.L.)
| | - Melissa L. Barron
- Department of Clinical Imaging, Faculty of Medicine and Health, The University of Sydney, Level 7-D18, Susan Wakil Health Building, Camperdown, NSW 2006, Australia; (N.B.); (T.L.); (P.C.B.); (M.L.B.); (S.J.L.)
| | - Sarah J. Lewis
- Department of Clinical Imaging, Faculty of Medicine and Health, The University of Sydney, Level 7-D18, Susan Wakil Health Building, Camperdown, NSW 2006, Australia; (N.B.); (T.L.); (P.C.B.); (M.L.B.); (S.J.L.)
| |
Collapse
|
5
|
Trieu PD(Y, Mello-Thoms CR, Barron ML, Lewis SJ. Look how far we have come: BREAST cancer detection education on the international stage. Front Oncol 2023; 12:1023714. [PMID: 36686760 PMCID: PMC9846523 DOI: 10.3389/fonc.2022.1023714] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/20/2022] [Accepted: 12/05/2022] [Indexed: 01/06/2023] Open
Abstract
The development of screening mammography over 30 years has remarkedly reduced breast cancer-associated mortality by 20%-30% through detection of small cancer lesions at early stages. Yet breast screening programmes may function differently in each nation depending on the incidence rate, national legislation, local health infrastructure and training opportunities including feedback on performance. Mammography has been the frontline breast cancer screening tool for several decades; however, it is estimated that there are 15% to 35% of cancers missed on screening which are owing to perceptual and decision-making errors by radiologists and other readers. Furthermore, mammography screening is not available in all countries and the increased speed in the number of new breast cancer cases among less developed countries exceeds that of the developed world in recent decades. Studies conducted through the BreastScreen Reader Assessment Strategy (BREAST) training tools for breast screening readers have documented benchmarking and significant variation in diagnostic performances in screening mammogram test sets in different countries. The performance of the radiologists from less well-established breast screening countries such as China, Mongolia and Vietnam were significant lower in detecting early-stage cancers than radiologists from developed countries such as Australia, USA, Singapore, Italy. Differences in breast features and cancer presentations, discrepancies in the level of experiences in reading screening mammograms, the availability of high-quality national breast screening program and breast image interpretation training courses between developed and less developed countries are likely to have impact on the variation of readers' performances. Hence dedicated education training programs with the ability to tailor to different reader cohorts and different population presentations are suggested to ameliorate challenges in exposure to a range of cancer cases and improve the interpretation skills of local radiologists. Findings from this review provide a good understanding of the radiologist' performances and their improvement using the education interventions, primarily the BREAST program, which has been deployed in a large range of developing and developed countries in the last decade. Self-testing and immediate feedback loops have been shown to have important implications for benchmarking and improving the diagnostic accuracy in radiology worldwide for better breast cancer control.
Collapse
Affiliation(s)
- Phuong Dung (Yun) Trieu
- Discipline of Medical Imaging Sciences, School of Health Sciences, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia
| | - Claudia R. Mello-Thoms
- Discipline of Medical Imaging Sciences, School of Health Sciences, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia
- Department of Radiology, Carver College of Medicine, University of Iowa, Iowa City, IA, United States
| | - Melissa L. Barron
- Discipline of Medical Imaging Sciences, School of Health Sciences, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia
| | - Sarah J. Lewis
- Discipline of Medical Imaging Sciences, School of Health Sciences, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia
| |
Collapse
|
6
|
Kataria B, Öman J, Sandborg M, Smedby Ö. Learning effects in visual grading assessment of model-based reconstruction algorithms in abdominal Computed Tomography. Eur J Radiol Open 2023; 10:100490. [PMID: 37207049 PMCID: PMC10189366 DOI: 10.1016/j.ejro.2023.100490] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/15/2023] [Revised: 04/06/2023] [Accepted: 05/01/2023] [Indexed: 05/21/2023] Open
Abstract
Objectives Images reconstructed with higher strengths of iterative reconstruction algorithms may impair radiologists' subjective perception and diagnostic performance due to changes in the amplitude of different spatial frequencies of noise. The aim of the present study was to ascertain if radiologists can learn to adapt to the unusual appearance of images produced by higher strengths of Advanced modeled iterative reconstruction algorithm (ADMIRE). Methods Two previously published studies evaluated the performance of ADMIRE in non-contrast and contrast-enhanced abdominal CT. Images from 25 (first material) and 50 (second material) patients, were reconstructed with ADMIRE strengths 3, 5 (AD3, AD5) and filtered back projection (FBP). Radiologists assessed the images using image criteria from the European guidelines for quality criteria in CT. To ascertain if there was a learning effect, new analyses of data from the two studies was performed by introducing a time variable in the mixed-effects ordinal logistic regression model. Results In both materials, a significant negative attitude to ADMIRE 5 at the beginning of the viewing was strengthened during the progress of the reviews for both liver parenchyma (first material: -0.70, p < 0.01, second material: -0.96, p < 0.001) and overall image quality (first material:-0.59, p < 0.05, second material::-1.26, p < 0.001). For ADMIRE 3, an early positive attitude for the algorithm was noted, with no significant change over time for all criteria except one (overall image quality), where a significant negative trend over time (-1.08, p < 0.001) was seen in the second material. Conclusions With progression of reviews in both materials, an increasing dislike for ADMIRE 5 images was apparent for two image criteria. In this time perspective (weeks or months), no learning effect towards accepting the algorithm could be demonstrated.
Collapse
Affiliation(s)
- Bharti Kataria
- Department of Radiology, Linköping University, Linköping, Sweden
- Department of Health, Medicine & Caring Sciences, Linköping University, Linköping, Sweden
- Center for Medical Image Science & Visualization (CMIV), Linköping University, Linköping, Sweden
- Correspondence to: County Council of Östergötland, DC, Department of Radiology, S-581 85 Linköping, Sweden.
| | - Jenny Öman
- Department of Radiology, Linköping University, Linköping, Sweden
- Department of Health, Medicine & Caring Sciences, Linköping University, Linköping, Sweden
| | - Michael Sandborg
- Department of Health, Medicine & Caring Sciences, Linköping University, Linköping, Sweden
- Center for Medical Image Science & Visualization (CMIV), Linköping University, Linköping, Sweden
- Department of Medical Physics, Linköping University, Linköping, Sweden
| | - Örjan Smedby
- Department of Biomedical Engineering and Health Systems (MTH), KTH Royal Institute of Technology, Stockholm, Sweden
| |
Collapse
|