1
|
Herington J, McCradden MD, Creel K, Boellaard R, Jones EC, Jha AK, Rahmim A, Scott PJH, Sunderland JJ, Wahl RL, Zuehlsdorff S, Saboury B. Ethical Considerations for Artificial Intelligence in Medical Imaging: Data Collection, Development, and Evaluation. J Nucl Med 2023; 64:1848-1854. [PMID: 37827839 PMCID: PMC10690124 DOI: 10.2967/jnumed.123.266080] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/28/2022] [Revised: 09/12/2023] [Indexed: 10/14/2023] Open
Abstract
The development of artificial intelligence (AI) within nuclear imaging involves several ethically fraught components at different stages of the machine learning pipeline, including during data collection, model training and validation, and clinical use. Drawing on the traditional principles of medical and research ethics, and highlighting the need to ensure health justice, the AI task force of the Society of Nuclear Medicine and Molecular Imaging has identified 4 major ethical risks: privacy of data subjects, data quality and model efficacy, fairness toward marginalized populations, and transparency of clinical performance. We provide preliminary recommendations to developers of AI-driven medical devices for mitigating the impact of these risks on patients and populations.
Collapse
Affiliation(s)
- Jonathan Herington
- Department of Health Humanities and Bioethics and Department of Philosophy, University of Rochester, Rochester, New York
| | - Melissa D McCradden
- Department of Bioethics, Hospital for Sick Children, Toronto and Dana Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
| | - Kathleen Creel
- Department of Philosophy and Religion and Khoury College of Computer Sciences, Northeastern University, Boston, Massachusetts
| | - Ronald Boellaard
- Department of Radiology and Nuclear Medicine, Cancer Centre Amsterdam, Amsterdam University Medical Centres, Amsterdam, The Netherlands
| | - Elizabeth C Jones
- Department of Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland
| | - Abhinav K Jha
- Department of Biomedical Engineering and Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri
| | - Arman Rahmim
- Departments of Radiology and Physics, University of British Columbia, Vancouver, British Columbia, Canada
| | - Peter J H Scott
- Department of Radiology, University of Michigan Medical School, Ann Arbor, Michigan
| | - John J Sunderland
- Departments of Radiology and Physics, University of Iowa, Iowa City, Iowa
| | - Richard L Wahl
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri; and
| | | | - Babak Saboury
- Department of Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland;
| |
Collapse
|
2
|
D’Arienzo M, Mezzenga E, Capotosti A, Bagni O, Filippi L, Capogni M, Indovina L, Sarnelli A. The Importance of Uncertainty Analysis and Traceable Measurements in Routine Quantitative 90Y-PET Molecular Radiotherapy: A Multicenter Experience. Pharmaceuticals (Basel) 2023; 16:1142. [PMID: 37631057 PMCID: PMC10459699 DOI: 10.3390/ph16081142] [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: 06/17/2023] [Revised: 08/01/2023] [Accepted: 08/03/2023] [Indexed: 08/27/2023] Open
Abstract
Molecular Radiation Therapy (MRT) is a valid therapeutic option for a wide range of malignancies, such as neuroendocrine tumors and liver cancers. In its practice, it is generally acknowledged that there is a need to evaluate the influence of different factors affecting the accuracy of dose estimates and to define the actions necessary to maintain treatment uncertainties at acceptable levels. The present study addresses the problem of uncertainty propagation in 90Y-PET quantification. We assessed the quantitative accuracy in reference conditions of three PET scanners (namely, Siemens Biograph mCT, Siemens Biograph mCT flow, and GE Discovery DST) available at three different Italian Nuclear Medicine centers. Specific aspects of uncertainty within the quantification chain have been addressed, including the uncertainty in the calibration procedure. A framework based on the Guide to the Expression of Uncertainty in Measurement (GUM) approach is proposed for modeling the uncertainty in the quantification processes, and ultimately, an estimation of the uncertainty achievable in clinical conditions is reported.
Collapse
Affiliation(s)
- Marco D’Arienzo
- Medical Physics Section, ASL Roma 6, Borgo Garibaldi 12, 00041 Rome, Italy;
- UniCamillus International Medical University, 00131 Rome, Italy
| | - Emilio Mezzenga
- Medical Physics Unit, IRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) “Dino Amadori”, 47014 Meldola, Italy; (E.M.); (A.S.)
| | - Amedeo Capotosti
- Fondazione Policlinico Universitario A. Gemelli IRCCS, 00168 Roma, Italy;
| | - Oreste Bagni
- Nuclear Medicine Department, Santa Maria Goretti Hospital, 04100 Latina, Italy; (O.B.); (L.F.)
| | - Luca Filippi
- Nuclear Medicine Department, Santa Maria Goretti Hospital, 04100 Latina, Italy; (O.B.); (L.F.)
| | - Marco Capogni
- ENEA, Italian National Institute of Ionizing Radiation Metrology, Via Anguillarese 301, 00123 Rome, Italy;
| | - Luca Indovina
- Fondazione Policlinico Universitario A. Gemelli IRCCS, 00168 Roma, Italy;
| | - Anna Sarnelli
- Medical Physics Unit, IRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) “Dino Amadori”, 47014 Meldola, Italy; (E.M.); (A.S.)
| |
Collapse
|
3
|
Dubal J, Arce P, South C, Florescu L. Cherenkov light emission in molecular radiation therapy of the thyroid and its application to dosimetry. BIOMEDICAL OPTICS EXPRESS 2022; 13:2431-2449. [PMID: 35519238 PMCID: PMC9045923 DOI: 10.1364/boe.448139] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 11/10/2021] [Revised: 03/07/2022] [Accepted: 03/11/2022] [Indexed: 06/14/2023]
Abstract
Numerical experiments based on Monte Carlo simulations and clinical CT data are performed to investigate the spatial and spectral characteristics of Cherenkov light emission and the relationship between Cherenkov light intensity and deposited dose in molecular radiotherapy of hyperthyroidism and papillary thyroid carcinoma. It is found that Cherenkov light is emitted mostly in the treatment volume, the spatial distribution of Cherenkov light at the surface of the patient presents high-value regions at locations that depend on the symmetry and location of the treatment volume, and the surface light in the near-infrared spectral region originates from the treatment site. The effect of inter-patient variability in the tissue optical parameters and radioisotope uptake on the linear relationship between the dose absorbed by the treatment volume and Cherenkov light intensity at the surface of the patient is investigated, and measurements of surface light intensity for which this effect is minimal are identified. The use of Cherenkov light measurements at the patient surface for molecular radiation therapy dosimetry is also addressed.
Collapse
Affiliation(s)
- Jigar Dubal
- Centre for Vision, Speech and Signal Processing, University of Surrey, GU2 7XH, United Kingdom
| | - Pedro Arce
- CIEMAT (Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas), 28040 Madrid, Spain
| | - Christopher South
- Department of Medical Physics, Royal Surrey County Hospital NHS Foundation Trust, Guildford GU2 7XX, United Kingdom
| | - Lucia Florescu
- Centre for Vision, Speech and Signal Processing, University of Surrey, GU2 7XH, United Kingdom
| |
Collapse
|
4
|
Sgouros G, Frey E, Du Y, Hobbs R, Bolch W. Imaging and dosimetry for alpha-particle emitter radiopharmaceutical therapy: improving radiopharmaceutical therapy by looking into the black box. Eur J Nucl Med Mol Imaging 2021; 49:18-29. [PMID: 34782911 DOI: 10.1007/s00259-021-05583-x] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/26/2021] [Accepted: 10/09/2021] [Indexed: 02/07/2023]
Abstract
Radiopharmaceutical therapy using α-particle emitting radionuclides (αRPT) is a novel treatment modality that delivers highly potent alpha-particles to cancer cells or their environment. We review the advantages and challenges of imaging and dosimetry in implementing αRPT for cancer patients.
Collapse
Affiliation(s)
| | - Eric Frey
- Johns Hopkins University, Baltimore, MD, USA
| | - Yong Du
- Johns Hopkins University, Baltimore, MD, USA
| | - Rob Hobbs
- Johns Hopkins University, Baltimore, MD, USA
| | | |
Collapse
|
5
|
Taprogge J, Wadsley J, Miles E, Flux GD. Recommendations for Multicentre Clinical Trials Involving Dosimetry for Molecular Radiotherapy. Clin Oncol (R Coll Radiol) 2021; 33:131-136. [PMID: 33342617 PMCID: PMC7818526 DOI: 10.1016/j.clon.2020.12.002] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/01/2020] [Revised: 11/06/2020] [Accepted: 12/02/2020] [Indexed: 11/17/2022]
Abstract
Multicentre clinical trials involving a dosimetry component are becoming more prevalent in molecular radiotherapy and are essential to generate the evidence to support individualised approaches to treatment planning and to ensure that sufficient patients are recruited to achieve the statistical significance required. Quality assurance programmes should be considered to support the standardisation required to achieve meaningful results. Trials should be designed to ensure that dosimetry results from image acquisition systems across centres are comparable by incorporating steps to standardise the methodologies used for the quantification of images and dosimetry. Furthermore, it is essential to assess the expertise and resources available at each participating site prior to trial commencement. A quality assurance plan should be drawn up and training provided if necessary. Standardisation of quantification and dosimetry methodologies used in a trial are essential to ensure that results from different centres may be collated. In addition, appropriate uncertainty analysis should be carried out to correct for differences in methodologies between centres. Recommendations are provided to support dosimetry studies based on the experience of several previous and ongoing multicentre trials.
Collapse
Affiliation(s)
- J Taprogge
- National Radiotherapy Trials Quality Assurance (RTTQA) Group, Joint Department of Physics, Royal Marsden NHSFT, Sutton, UK; The Institute of Cancer Research, London, UK.
| | | | - E Miles
- National Radiotherapy Trials Quality Assurance (RTTQA) Group, Mount Vernon Cancer Centre, East and North Hertfordshire NHS Trust, Northwood, UK
| | - G D Flux
- The Institute of Cancer Research, London, UK; Joint Department of Physics, Royal Marsden NHSFT, Sutton, UK
| |
Collapse
|
6
|
Sanghera B, Fenwick A, Lowe G, Sullivan K, Wong WL. Radionuclide calibrator intercomparison study of clinical PET centres in England to a single traceable 68Ge syringe source. Nucl Med Commun 2020; 41:965-976. [PMID: 32796486 DOI: 10.1097/mnm.0000000000001241] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/27/2022]
Abstract
OBJECTIVES The aim of this study was to characterize national variation in radionuclide calibrator activity response to a single National Institute of Standards and Technology (NIST) traceable reference Ge source used as a surrogate for F at clinical PET centres in England using National Physical Laboratory approved techniques. METHODS Readings from 20 instruments at 13 centres using local F and Ge factor settings were recorded with the source located in vial and syringe positions. Ten repeat measurements were conducted to investigate repeatability using % coefficient of variability (COV). Comparison ratios to investigate accuracy were made between calibrator responses and decay-corrected NISTref reference activity for syringe and vial position measurements. RESULTS The maximum %COV was 0.79%, while 90, 95 and 80% of calibrators conformed to 5% accuracy for F syringe, Ge syringe and Ge vial position readings, respectively. We revealed a trend towards reduced bias in measurements using Veenstra devices for F and using Capintec devices for Ge factor settings. CONCLUSIONS This study demonstrated good repeatability in local device measurements. In total, 70% of English calibrators tested and 88% of all measurements performed achieved 5% accuracy. While statistically significant bias was exhibited between different vendor equipment dependent upon radioisotope selected, our study recommends regular traceability checks for optimum instrument performance conducted within National Metrology Institutes guidelines.
Collapse
Affiliation(s)
- Bal Sanghera
- Paul Strickland Scanner Centre, Mount Vernon Hospital, Northwood
| | | | - Gerry Lowe
- Cancer Centre, Mount Vernon Hospital, Northwood
| | - Keith Sullivan
- Health Research Methods Unit, University of Hertfordshire, UK
| | - Wai-Lup Wong
- Paul Strickland Scanner Centre, Mount Vernon Hospital, Northwood
| |
Collapse
|
7
|
Taprogge J, Leek F, Flux GD. Physics aspects of setting up a multicenter clinical trial involving internal dosimetry of radioiodine treatment of differentiated thyroid cancer. THE QUARTERLY JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING : OFFICIAL PUBLICATION OF THE ITALIAN ASSOCIATION OF NUCLEAR MEDICINE (AIMN) [AND] THE INTERNATIONAL ASSOCIATION OF RADIOPHARMACOLOGY (IAR), [AND] SECTION OF THE SOCIETY OF... 2019; 63:271-277. [PMID: 31315346 DOI: 10.23736/s1824-4785.19.03202-3] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 02/11/2024]
Abstract
The field of molecular radiotherapy is expanding rapidly, with the advent of many new radiotherapeutics for the treatment of common as well as for rare cancers. Treatment outcome is dependent on the absorbed doses delivered to target volumes and to healthy organs-at-risk, which are shown to vary widely from fixed administrations of activity. There have been significant developments in quantitative imaging and internal dosimetry in recent years, although clinical implementation of these methods has been slow in comparison with external beam radiotherapy, partly due to there being relatively few patients treated at single centers. Multicenter clinical trials are therefore essential to acquire the data required to ensure best practice and to develop the personalized treatment planning that this area is well suited to, due to the unrivalled opportunity to image the therapeutic drug in vivo. Initial preparation for such trials requires a significant effort in terms of resources and trial design. Imaging systems in participating centers must be characterized and set up for quantitative imaging to allow for collation of data. Data transfer for centralized processing is usually necessary but is hindered in some cases by data protection regulations and local logistics. Recent multicenter clinical trials involving radioiodine therapy have begun to establish the procedures necessary for quantitative SPECT imaging in a multicenter setting using standard and anthropomorphic phantoms. The establishment of national and international multicenter imaging and dosimetry networks will provide frameworks to develop and harmonize best practice with existing therapeutic procedures and to ensure rapid and optimized clinical implementation of new radiotherapeutics across all centers of excellence that offer molecular radiotherapy. This will promote networks and collaborations that can provide a sound basis for further developments and will ensure that nuclear medicine maintains a key role in future developments.
Collapse
Affiliation(s)
- Jan Taprogge
- Joint Department of Physics, Royal Marsden NHS Foundation Trust, Sutton, UK -
- The Institute of Cancer Research, London, UK -
| | - Francesca Leek
- Joint Department of Physics, Royal Marsden NHS Foundation Trust, Sutton, UK
- The Institute of Cancer Research, London, UK
| | - Glenn D Flux
- Joint Department of Physics, Royal Marsden NHS Foundation Trust, Sutton, UK
- The Institute of Cancer Research, London, UK
| |
Collapse
|
8
|
Denis-Bacelar AM, Chittenden SJ, McCready VR, Divoli A, Dearnaley DP, O’Sullivan JM, Johnson B, Flux GD. Bone lesion absorbed dose profiles in patients with metastatic prostate cancer treated with molecular radiotherapy. Br J Radiol 2018; 91:20170795. [PMID: 29293372 PMCID: PMC5966004 DOI: 10.1259/bjr.20170795] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/16/2017] [Revised: 12/06/2017] [Accepted: 12/21/2017] [Indexed: 11/05/2022] Open
Abstract
OBJECTIVE The aim of this study was to calculate the range of absorbed doses that could potentially be delivered by a variety of radiopharmaceuticals and typical fixed administered activities used for bone pain palliation in a cohort of patients with metastatic castration-resistant prostate cancer (mCRPC). The methodology for the extrapolation of the biodistribution, pharmacokinetics and absorbed doses from a given to an alternative radiopharmaceutical is presented. METHODS Sequential single photon emission CT images from 22 patients treated with 5 GBq of 186Re-HEDP were used to extrapolate the time-activity curves for various radiopharmaceuticals. Cumulated activity distributions for the delivered and extrapolated treatment plans were converted into absorbed dose distributions using the convolution dosimetry method. The lesion absorbed doses obtained for the different treatments were compared using the patient population distributions and cumulative dose-volume histograms. RESULTS The median lesion absorbed doses across the patient cohort ranged from 2.7 Gy (range: 0.6-11.8 Gy) for 1100 MBq of 166Ho-DOTMP to 21.8 Gy (range: 4.5-117.6 Gy) for 150 MBq of 89Sr-dichloride. 32P-Na3PO4, 153Sm-EDTMP, 166Ho-DOTMP, 177Lu-EDTMP and 188Re-HEDP would have delivered 41, 32, 85, 20 and 64% lower absorbed doses, for the typical administered activities as compared to 186Re-HEDP, respectively, whilst 89Sr-dichloride would have delivered 25% higher absorbed doses. CONCLUSION For the patient cohort studied, a wide range of absorbed doses would have been delivered for typical administration protocols in mCRPC. The methodology presented has potential use for emerging theragnostic agents. Advances in knowledge: The same patient cohort can receive a range of lesion absorbed doses from typical molecular radiotherapy treatments for patients with metastatic prostate cancer, highlighting the need to establish absorbed dose response relationships and to treat patients according to absorbed dose instead of using fixed administered activities.
Collapse
Affiliation(s)
| | - Sarah J Chittenden
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden Hospital NHS Foundation Trust, London, UK
| | - V Ralph McCready
- Department of Nuclear Medicine, Royal Sussex County Hospital, Brighton, UK
| | - Antigoni Divoli
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden Hospital NHS Foundation Trust, London, UK
| | - David P Dearnaley
- Division of Radiotherapy and Imaging, The Institute of Cancer Research and The Royal Marsden Hospital NHS Foundation Trust, London, UK
| | - Joe M O’Sullivan
- Centre for Cancer Research and Cell Biology, Queen’s University Belfast, Belfast, UK
| | - Bernadette Johnson
- Division of Radiotherapy and Imaging, The Institute of Cancer Research and The Royal Marsden Hospital NHS Foundation Trust, London, UK
| | - Glenn D Flux
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden Hospital NHS Foundation Trust, London, UK
| |
Collapse
|