Social Asymmetry, Artificial Intelligence and the Medical Imaging Landscape.

Social and health care equity and justice should be prioritized by the mantra of medicine, first do no harm. Despite highly motivated national and global health strategies, there remains significant health care inequity. Intrinsic and extrinsic factors, including a number of biases, are key drivers of ongoing health inequity including equity of access and opportunity for nuclear medicine and radiology services. There is a substantial gap in the global practice of nuclear medicine in particular, but also radiology, between developed health economies and those considered developing or undeveloped. At a local level, even in developed health economies, there can be a significant disparity between health services, including medical imaging, between communities based on socioeconomic, cultural or geographic differences. Artificial intelligence (AI) has the potential to either widen the health inequity divide or substantially reduce it. Distributed generally, AI technology could be used to overcome geographic boundaries to health care, thus bringing general and specialist care into underserved communities. However, should AI technology be limited to localities already enjoying ample healthcare access and direct access to health infrastructure, like radiology and nuclear medicine, it could then accentuate the gap. There are a number of challenges across the AI pipeline that need careful attention to ensure beneficence over maleficence. Fully realized, AI augmented health care could be crafted as an integral part of the broader strategy convergence on local, national and global health equity. The applications of AI in nuclear medicine and radiology could emerge as a powerful tool in social and health equity.

Chemotherapy Response Assessment by FDG-PET-CT in Early-stage Classical Hodgkin Lymphoma: Moving Beyond the Five-Point Deauville Score.

PURPOSE: In early-stage classical Hodgkin lymphoma, fluorodeoxyglucose positron emission tomography (PET)-computed tomography (CT) scans are performed routinely after chemotherapy, and the 5-point Deauville score is used to report the disease response. We hypothesized that other PET-CT parameters, considered in combination with Deauville score, would improve risk stratification. METHODS AND MATERIALS: Patients treated for stage I to II Hodgkin lymphoma from 2003 to 2013, who were aged ≥18 years and had analyzable PET-CT scans performed before and after chemotherapy, were eligible. The soft tissue volume (STV), maximum standardized uptake value, metabolic tumor volume, and total lesion glycolysis were recorded from the PET-CT scans before and after chemotherapy. Reductions were defined as 1 - (final PET-CT value)/(corresponding initial PET-CT value). The primary endpoint was freedom from progression (FFP). RESULTS: For 202 patients treated with chemotherapy with or without radiation therapy, the 5-year FFP was 89% (95% confidence interval 85%-93%). All PET-CT parameters were strongly associated with the Deauville score (P<.001) and FFP (P<.0001) on univariate analysis. The Deauville score was highly predictive of FFP (C-index 0.89) but was less discriminating in the Deauville 1 to 4 subset (C-index 0.67). Therefore, we aimed to identify PET-CT parameters that would improve risk stratification for this subgroup (n=187). STV reduction was predictive of outcome (C-index 0.71) and was dichotomized with an optimal cutoff of 0.65 (65% reduction in STV). A model incorporating the Deauville score and STV reduction predicted FFP more accurately than either measurement alone in the Deauville 1 to 4 subset (C-index 0.83). The improvement in predictive accuracy of this composite measure compared with the Deauville score alone met statistical significance (P=.045). CONCLUSIONS: The relative reduction in tumor size is an independent predictor of outcome. Combined with the Deauville score, it might improve risk stratification and contribute to response-adapted individualization of therapy.

Assessment of Sequential PET/MRI in Comparison With PET/CT of Pediatric Lymphoma: A Prospective Study.

OBJECTIVE: The objective of our study was to compare the diagnostic performance of sequential (18)F-FDG PET/MRI (PET/MRI) and (18)F-FDG PET/CT (PET/CT) in a pediatric cohort with lymphoma for lesion detection, lesion classification, and disease staging; quantification of FDG uptake; and radiation dose. SUBJECTS AND METHODS: For this prospective study of 25 pediatric patients with lymphoma, 40 PET/CT and PET/MRI examinations were performed after a single-injection dual-time-point imaging protocol. Lesions detected, lesion classification, Ann Arbor stage, and radiation dose were tabulated for each examination, and statistical evaluations were performed to compare the modalities. Quantification of standardized uptake values (SUVs) was performed for all lesions. All available examinations and clinical history were used as the reference standard. RESULTS: No statistically significant differences between PET/MRI and PET/CT were observed in lesion detection rates, lesion classification, or Ann Arbor staging. Fifty-four regions of focal uptake were observed on PET/MRI compared with 55 on PET/CT. Both modalities accurately classified 82% of the lesions relative to the reference standard. Disease staging based on PET/MRI was correct for 35 of the 40 studies, and disease staging based on PET/CT was correct for 35 of the 40 studies; there was substantial agreement between the modalities for disease staging (κ = 0.684; p < 0.001). PET SUVs were strongly correlated between PET/CT and PET/MRI (ρ > 0.72), although PET/MRI showed systematically lower SUV measurements. PET/MRI offered an average 45% reduction in radiation dose relative to PET/CT. CONCLUSION: In a pediatric cohort with lymphoma, sequential PET/MRI showed lesion detection, lesion classification, and Ann Arbor staging comparable to PET/CT. PET/MRI quantification of FDG uptake strongly correlated with PET/CT, but the SUVs were not interchangeable. PET/MRI significantly reduced radiation exposure and is a promising new alternative in the care of pediatric lymphoma patients.

Qualitative FDG PET Image Assessment Using Automated Three-Segment MR Attenuation Correction Versus CT Attenuation Correction in a Tertiary Pediatric Hospital: A Prospective Study.

OBJECTIVE: The purpose of this study was to systematically evaluate the diagnostic quality of (18)F-FDG PET images generated using MR attenuation correction (MRAC) compared with those images generated using CT attenuation correction (CTAC) in a pediatric population. SUBJECTS AND METHODS: Forty-two patients (mean age, 12.8 years; percentage who were male, 57%) who were referred for 62 indicated whole-body PET/CT studies were prospectively recruited to undergo PET/MRI examinations during the same clinic visit in which PET/CT was performed. MRAC was performed using an automatic three-segment model. Three nuclear radiologists scored the diagnostic quality of the PET images generated by MRAC and CTAC using a Likert scale (range of scores, 1-5). Images graded with a score of 1-3 were considered clinically unacceptable, whereas images with a score of 4-5 were considered clinically acceptable. A Wilcoxon signed-rank test was used to compare differences in the grading of PET/MRI and PET/CT images. The Fisher exact test was used to evaluate potential differences in clinically acceptable image quality and the presence of artifact. Fleiss kappa statistics were used to examine interobserver agreement. RESULTS: There was no statistically significant difference in the proportion of PET images generated with MRAC and CTAC for which image quality was considered clinically acceptable. A total of 3.9% of PET assessments generated with MRAC were of unacceptable image quality, compared with 2.2% of PET images generated with CTAC. Two of the three radiologists who reviewed the PET images reported the presence of artifacts more often on MRAC-derived images, and they graded the mean quality of these images 0.48 and 0.29 points lower on the 5-point Likert scale than they graded the mean quality of CTAC-derived images (p < 0.0001). Interobserver agreement was fair (κ = 0.39). CONCLUSION: The diagnostic quality of PET images obtained from a pediatric population with the use of an automatic three-segmentation MRAC method was comparable to that of PET images obtained with the use of CTAC.

Positron emission tomography-computed tomography reporting in radiation therapy planning and response assessment.

Quality reporting in radiology and nuclear medicine has been described in 8 Cs: correctness, completeness, consistency, communication, clarity, confidence, concision, and consultation. The incorporation of all of these elements is necessary for the generation of high-quality, effective reports. Positron emission tomography-computed tomography presents a particular challenge because of the large amount of information contained in a single study, both anatomic and metabolic. In this article, the elements of good reporting will be reviewed with specific attention to positron emission tomography-computed tomography, including specific case examples that demonstrate effective and ineffective reporting styles. The principles of reporting will then be applied specifically to reporting for radiotherapy planning and reporting for response assessment.

Response assessment of aggressive non-Hodgkin's lymphoma by integrated International Workshop Criteria and fluorine-18-fluorodeoxyglucose positron emission tomography.

PURPOSE: To determine whether a response classification based on integration of fluorine-18-fluorodeoxyglucose positron emission tomography (FDG-PET) into the International Workshop Criteria (IWC) provides a more accurate response assessment than IWC alone in patients with non-Hodgkin's lymphoma (NHL). PATIENTS AND METHODS: Fifty-four patients with aggressive NHL who underwent FDG-PET and computed tomography 1 to 16 weeks after four to eight cycles of chemotherapy with cyclophosphamide, doxorubicin, vincristine, and prednisone were assessed for complete response (CR), unconfirmed CR (CRu), partial response (PR), stable disease (SD), and progressive disease (PD) by the IWC and by integrated IWC and FDG-PET (IWC+PET). Progression-free survival (PFS) was also compared between IWC- and IWC+PET-assigned response designations. RESULTS: By IWC, 17 patients had a CR, seven had a CRu, 19 had a PR, nine had SD, and two had PD. In comparison, by IWC+PET, 35 patients had a CR, 12 had a PR, six had SD, one had PD, and zero had a CRu. In separate multivariate models, PFS was significantly shorter in patients with PR than in those with a CR using IWC (hazard ratio [HR], 8.9; P = .021) or IWC+PET (HR, 29.7; P = .0003). However, when the two classifications were included in the same multivariate model, only IWC+PET was a statistically significant independent predictor for PFS (P = .008 v P = .72 for IWC). In addition, when patients with a PR by IWC and a CR by IWC+PET were compared with those with a CR by IWC and a CR by IWC+PET, there was no significant difference in PFS (HR, 1.6; P = .72), indicating that IWC+PET identified a subset of IWC-PR patients with a more favorable prognosis. CONCLUSION: Compared with IWC, the IWC+PET-based assessment provides a more accurate response classification in patients with aggressive NHL.

The role of F-18 FDG positron emission tomography in preoperative assessment of the liver in patients being considered for curative resection of hepatic metastases from colorectal cancer.

PURPOSE: The authors' goal was to determine the sensitivity and specificity of F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) for identifying patients with hepatic metastases from colorectal cancer and the accuracy of PET for determining the number and distribution of lesions within the liver. Intraoperative sonography and surgical inspection and palpation were used as the reference standard. METHODS: Twenty-three patients being evaluated for surgical resection of hepatic metastases from colorectal carcinoma underwent FDG PET before operation. Findings of the PET studies were reviewed in a blinded, retrospective manner, with the results compared with the findings of intraoperative sonography and surgical exploration. Lesions of all sizes were considered in the analysis. RESULTS: The FDG-PET results were positive in 21 of the 22 patients ultimately found to have metastatic disease to the liver, and they were negative in the single patient without metastases. Therefore, for identification of patients with hepatic metastatic disease, PET has a sensitivity of 95% and a specificity of 100%. In all, 48 metastatic lesions were identified in these patients, of which 38 (79%) were identified on PET images. The probability of lesion detection by PET was directly correlated with lesion size (P < 0.01). The assessment of lobar disease distribution in the liver was discordant between PET and surgery in 3 of 23 (13%) patients. CONCLUSIONS: In patients being evaluated for potential curative resection of hepatic metastases from colorectal cancer, FDG PET is accurate for the identification of the presence or absence of metastatic disease to the liver. However, detection of individual lesions depends on their size, and determination of lesion number and distribution within the liver is more accurately accomplished with intraoperative sonography.