Virva Saunavaara
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Virva Saunavaara has a total of 327 co-authors. Here are the top co-authors by number of shared publications:
- Jarmo Teuho
Turku PET Centre, University of Turku and Turku University Hospital, Turku, Finland
14 shared publications - Juhani Knuuti
Turku PET Centre, University of Turku, Turku, Finland
11 shared publications Mika Teräs
11 shared publications- Jarna Christina Hannukainen
Turku PET Centre, University of Turku, Turku, Finland
10 shared publications Riitta Parkkola
Department of Radiology University of Turku Turku Finland
9 shared publications- Anne Roivainen
University of Turku and Turku University Hospital
8 shared publications
Latest Publications
- Open Access
- 26 June 2026
- Published by Springer Nature in European Journal of Nuclear Medicine and Molecular Imaging
- Vol. 53 (12) , 6837-6848
Comparison of [18F]DCFPyL and [18F]FAPI-74 PET/CT imaging in metastatic prostate cancer
Purpose: To compare prostate-specific membrane antigen (PSMA) and fibroblast activation-protein inhibitor (FAPI) targeted PET/CT imaging to assess possible theranostic combination strategies of both tracers in metastatic prostate cancer (mPC). Methods: This prospective study enrolled patients with mPC who underwent long-axial field-of-view whole-body PET/CT with 18 F-DCFPyL and 18 F-FAPI-74 within two weeks. Fifteen patients had hormone sensitive (mHSPC) and 24 castration-resistant (mCRPC) disease. Scans were reviewed according to predefined criteria, with up to five metastases per anatomical region evaluated. Lesions were categorized in four groups as skeletal, pelvic nodal, extrapelvic nodal, or visceral. SUVmax and volumetric parameters were measured for each lesion and at the patient level. Results: A total of 1365 lesions were identified in 39 patients. 18 F-DCFPyL PET/CT detected 1351 lesions (99%). In contrast, 18 F–FAPI-74 identified only 788 (58%) lesions in the PSMA-guided analysis and 523 (38%) lesions in the independent analysis. Median lesion SUVmax was higher for PSMA (24) than for FAPI (5.4) (p < 0.001). Visceral metastases constituted 3% of lesions with both tracers and tended to be more often FAPI-avid (64%) compared to other lesion groups. Only 14 (1%) of lesions were FAPI-positive and PSMA-negative. In whole-tumor burden analysis, median tracer-positive tumor volume on 18 F-DCFPyL PET/CT was 85 mL and on 18 F-FAPI-74 PET/CT only 10 mL (p < 0.001). Conclusion: In mPC, PSMA PET/CT remains the preferred molecular imaging method. Our findings suggest only a limited role of FAPI radioligand therapy. FAPI theranostics should focus on patients with visceral metastases and specific histological subtypes beyond acinar adenocarcinoma. Clinical Trial Registration: EudraCT: 2022–003788-13. Registered 21 September 2023
- 26 June 2026
- Published by Springer Nature in European Journal of Nuclear Medicine and Molecular Imaging
- Vol. 53 (12) , 6837-6848
Open AccessComparison of [18F]DCFPyL and [18F]FAPI-74 PET/CT imaging in metastatic prostate cancer
Purpose: To compare prostate-specific membrane antigen (PSMA) and fibroblast activation-protein inhibitor (FAPI) targeted PET/CT imaging to assess possible theranostic combination strategies of both tracers in metastatic prostate cancer (mPC). Methods: This prospective study enrolled patients with mPC who underwent long-axial field-of-view whole-body PET/CT with 18 F-DCFPyL and 18 F-FAPI-74 within two weeks. Fifteen patients had hormone sensitive (mHSPC) and 24 castration-resistant (mCRPC) disease. Scans were reviewed according to predefined criteria, with up to five metastases per anatomical region evaluated. Lesions were categorized in four groups as skeletal, pelvic nodal, extrapelvic nodal, or visceral. SUVmax and volumetric parameters were measured for each lesion and at the patient level. Results: A total of 1365 lesions were identified in 39 patients. 18 F-DCFPyL PET/CT detected 1351 lesions (99%). In contrast, 18 F–FAPI-74 identified only 788 (58%) lesions in the PSMA-guided analysis and 523 (38%) lesions in the independent analysis. Median lesion SUVmax was higher for PSMA (24) than for FAPI (5.4) (p < 0.001). Visceral metastases constituted 3% of lesions with both tracers and tended to be more often FAPI-avid (64%) compared to other lesion groups. Only 14 (1%) of lesions were FAPI-positive and PSMA-negative. In whole-tumor burden analysis, median tracer-positive tumor volume on 18 F-DCFPyL PET/CT was 85 mL and on 18 F-FAPI-74 PET/CT only 10 mL (p < 0.001). Conclusion: In mPC, PSMA PET/CT remains the preferred molecular imaging method. Our findings suggest only a limited role of FAPI radioligand therapy. FAPI theranostics should focus on patients with visceral metastases and specific histological subtypes beyond acinar adenocarcinoma. Clinical Trial Registration: EudraCT: 2022–003788-13. Registered 21 September 2023
- Open Access
- 6 August 2025
- Published by Elsevier in JHEP Reports
- Vol. 7 (11) , 101535
Mildly elevated liver lipid content is characterised by reduced insulin sensitivity
- 6 August 2025
- Published by Elsevier in JHEP Reports
- Vol. 7 (11) , 101535
Open AccessMildly elevated liver lipid content is characterised by reduced insulin sensitivity
- Open Access
- 23 August 2023
- Published by Springer Nature in BMC Pediatrics
- Vol. 23 (1) , 1-9
Web-based follow-up tool (ePIPARI) of preterm infants—study protocol for feasibility and performance
Background: Preterm infants have a risk of health and developmental problems emerging after discharge. This indicates the need for a comprehensive follow-up to enable early identification of these problems. In this paper, we introduce a follow-up tool “ePIPARI – web-based follow-up for preterm infants”. Our future aim is to investigate whether ePIPARI is a feasible tool in the follow-up of preterm infants and whether it can identify children and parents in need of clinical interventions. Methods: ePIPARI includes eight assessment points (at term age and at 1, 2, 4, 8, 12, 18, and 24 months of corrected age) when the child´s health and growth, eating and feeding, neurodevelopment, and parental well-being are evaluated. ePIPARI consists of several widely used, standardized questionnaires, in addition to questions typically presented to parents in clinical follow-up visits. It also provides video guidance and written information about age-appropriate neurodevelopment for the parents. Parents of children born before 34 weeks of gestation during years 2019–2022 are being invited to participate in the ePIPARI study, in which web-based follow-up with ePIPARI is compared to clinical follow-up. In addition, the parents of children born before 32 weeks of gestation, who reached the corrected age of two years during 2019–2021 were invited to participate for the assessment point of 24 months of ePIPARI. The parents are asked to fill in the online questionnaires two weeks prior to each clinical follow-up visit. Discussion: The web-based tool, ePIPARI, was developed to acquire a sensitive and specific tool to detect infants and parents in need of further support and clinical interventions. This tool could allow individualized adjustments of the frequency and content of the clinical visits. Trial registration: ClinicalTrials.cov, NCT05238168. Registered 11 April 2022 – Retrospectively registered.
- 23 August 2023
- Published by Springer Nature in BMC Pediatrics
- Vol. 23 (1) , 1-9
Open AccessWeb-based follow-up tool (ePIPARI) of preterm infants—study protocol for feasibility and performance
Background: Preterm infants have a risk of health and developmental problems emerging after discharge. This indicates the need for a comprehensive follow-up to enable early identification of these problems. In this paper, we introduce a follow-up tool “ePIPARI – web-based follow-up for preterm infants”. Our future aim is to investigate whether ePIPARI is a feasible tool in the follow-up of preterm infants and whether it can identify children and parents in need of clinical interventions. Methods: ePIPARI includes eight assessment points (at term age and at 1, 2, 4, 8, 12, 18, and 24 months of corrected age) when the child´s health and growth, eating and feeding, neurodevelopment, and parental well-being are evaluated. ePIPARI consists of several widely used, standardized questionnaires, in addition to questions typically presented to parents in clinical follow-up visits. It also provides video guidance and written information about age-appropriate neurodevelopment for the parents. Parents of children born before 34 weeks of gestation during years 2019–2022 are being invited to participate in the ePIPARI study, in which web-based follow-up with ePIPARI is compared to clinical follow-up. In addition, the parents of children born before 32 weeks of gestation, who reached the corrected age of two years during 2019–2021 were invited to participate for the assessment point of 24 months of ePIPARI. The parents are asked to fill in the online questionnaires two weeks prior to each clinical follow-up visit. Discussion: The web-based tool, ePIPARI, was developed to acquire a sensitive and specific tool to detect infants and parents in need of further support and clinical interventions. This tool could allow individualized adjustments of the frequency and content of the clinical visits. Trial registration: ClinicalTrials.cov, NCT05238168. Registered 11 April 2022 – Retrospectively registered.
- Open Access
- 25 April 2023
- Published by European Publishing in Tobacco Prevention & Cessation
- Vol. 9 (Supplement)
Maternal smoking during pregnancy negatively affects brain volumes proportional to intracranial volume in adolescents born very preterm
Introduction: Maternal smoking during pregnancy has been shown to associate with smaller frontal lobe and cerebellar volumes in brain magnetic resonance imaging (MRI) at term age in very preterm infants. The aim of this study was to examine the effect of maternal smoking during pregnancy on...
- 25 April 2023
- Published by European Publishing in Tobacco Prevention & Cessation
- Vol. 9 (Supplement)
Open AccessMaternal smoking during pregnancy negatively affects brain volumes proportional to intracranial volume in adolescents born very preterm
Introduction: Maternal smoking during pregnancy has been shown to associate with smaller frontal lobe and cerebellar volumes in brain magnetic resonance imaging (MRI) at term age in very preterm infants. The aim of this study was to examine the effect of maternal smoking during pregnancy on...
- Open Access
- 24 August 2022
- Published by Wiley in Contrast Media & Molecular Imaging
- Vol. 2022 (1) , 8676787
Diagnostic Accuracy of 18F‐FDG‐PET/CT and 18F‐FDG‐PET/MRI in Detecting Locoregional Recurrence of HNSCC 12 Weeks after the End of Chemoradiotherapy: Single‐Center Experience with PET/MRI
Purpose. In head and neck squamous cell carcinoma (HNSCC), the early diagnosis and efficient detection of recurrences and/or residual tumor after treatment play a very important role in patient’s prognosis. Positron emission tomography (PET) using 2‐deoxy‐2‐18F‐fluoro‐D‐glucose (18F‐FDG) has become an established method for the diagnosis of suspected recurrence in head and neck carcinomas. In particular, integrated PET/MRI imaging that provides optimal soft tissue contrast and less dental implant artifacts compared to PET/CT is an intriguing technique for the follow‐up imaging of HNSCC patients. The aim of this study was to evaluate the benefit of PET/MRI compared to PET/CT in post‐treatment follow‐up imaging of HNSCC patients. Methods. This retrospective observational cohort study consists of 104 patients from our center with histologically confirmed HNSCC. All patients received chemoradiotherapy (CRT) and underwent 18F‐FDG‐PET/CT (n = 52) or 18F‐FDG‐PET/MRI (n = 52) scan 12 weeks after the end of treatment. Image analysis was performed by two independent readers according to a five‐point Likert scale analysis. Results. PET/MRI was more sensitive (1.00 vs. 0.77) than PET/CT in the detection of locoregional recurrence. PET/MRI also had better negative (1.00 vs. 0.87) predictive values. AUCs for PET/MRI and PET/CT on patient‐based analysis were 0.997 (95% CI 0.989–1.000) and 0.890 (95% CI 0.806–0.974), respectively. The comparison of sensitivity, AUCs, and negative predictive values revealed a statistically significant difference, p < 0.05. In PET/CT, false‐negative and positive findings were observed in the more advanced disease stages, where PET/MRI performed better. Also, false‐negative findings were located in the oropharyngeal, laryngeal, and nasopharyngeal regions, where PET/MRI made no false‐negative interpretations. Conclusion. Based on these results, PET/MRI might be considered the modality of choice in detecting locoregional recurrence in HNSCC patients, especially in the more advanced stages in the oral cavity, larynx, or nasopharynx.
- 24 August 2022
- Published by Wiley in Contrast Media & Molecular Imaging
- Vol. 2022 (1) , 8676787
Open AccessDiagnostic Accuracy of 18F‐FDG‐PET/CT and 18F‐FDG‐PET/MRI in Detecting Locoregional Recurrence of HNSCC 12 Weeks after the End of Chemoradiotherapy: Single‐Center Experience with PET/MRI
Purpose. In head and neck squamous cell carcinoma (HNSCC), the early diagnosis and efficient detection of recurrences and/or residual tumor after treatment play a very important role in patient’s prognosis. Positron emission tomography (PET) using 2‐deoxy‐2‐18F‐fluoro‐D‐glucose (18F‐FDG) has become an established method for the diagnosis of suspected recurrence in head and neck carcinomas. In particular, integrated PET/MRI imaging that provides optimal soft tissue contrast and less dental implant artifacts compared to PET/CT is an intriguing technique for the follow‐up imaging of HNSCC patients. The aim of this study was to evaluate the benefit of PET/MRI compared to PET/CT in post‐treatment follow‐up imaging of HNSCC patients. Methods. This retrospective observational cohort study consists of 104 patients from our center with histologically confirmed HNSCC. All patients received chemoradiotherapy (CRT) and underwent 18F‐FDG‐PET/CT (n = 52) or 18F‐FDG‐PET/MRI (n = 52) scan 12 weeks after the end of treatment. Image analysis was performed by two independent readers according to a five‐point Likert scale analysis. Results. PET/MRI was more sensitive (1.00 vs. 0.77) than PET/CT in the detection of locoregional recurrence. PET/MRI also had better negative (1.00 vs. 0.87) predictive values. AUCs for PET/MRI and PET/CT on patient‐based analysis were 0.997 (95% CI 0.989–1.000) and 0.890 (95% CI 0.806–0.974), respectively. The comparison of sensitivity, AUCs, and negative predictive values revealed a statistically significant difference, p < 0.05. In PET/CT, false‐negative and positive findings were observed in the more advanced disease stages, where PET/MRI performed better. Also, false‐negative findings were located in the oropharyngeal, laryngeal, and nasopharyngeal regions, where PET/MRI made no false‐negative interpretations. Conclusion. Based on these results, PET/MRI might be considered the modality of choice in detecting locoregional recurrence in HNSCC patients, especially in the more advanced stages in the oral cavity, larynx, or nasopharynx.
- Open Access
- 1 July 2022
- Published by Springer Nature in Pediatric Radiology
- Vol. 53 (1) , 94-103
Microstructural alterations in association tracts and language abilities in schoolchildren born very preterm and with poor fetal growth
Background: Prematurity and perinatal risk factors may influence white matter microstructure. In turn, these maturational changes may influence language development in this high-risk population of children. Objective: To evaluate differences in the microstructure of association tracts between preterm and term children and between preterm children with appropriate growth and those with fetal growth restriction and to study whether the diffusion tensor metrics of these tracts correlate with language abilities in schoolchildren with no severe neurological impairment. Materials and methods: This study prospectively followed 56 very preterm children (mean gestational age: 28.7 weeks) and 21 age- and gender-matched term children who underwent diffusion tensor imaging at a mean age of 9 years. We used automated probabilistic tractography and measured fractional anisotropy in seven bilateral association tracts known to belong to the white matter language network. Both groups participated in language assessment using five standardised tests at the same age. Results: Preterm children had lower fractional anisotropy in the right superior longitudinal fasciculus 1 compared to term children (P < 0.05). Preterm children with fetal growth restriction had lower fractional anisotropy in the left inferior longitudinal fasciculus compared to preterm children with appropriate fetal growth (P < 0.05). Fractional anisotropy in three dorsal tracts and in two dorsal and one ventral tract had a positive correlation with language assessments among preterm children and preterm children with fetal growth restriction, respectively (P < 0.05). Conclusion: There were some microstructural differences in language-related tracts between preterm and term children and between preterm children with appropriate and those with restricted fetal growth. Children with better language abilities had a higher fractional anisotropy in distinct white matter tracts.
- 1 July 2022
- Published by Springer Nature in Pediatric Radiology
- Vol. 53 (1) , 94-103
Open AccessMicrostructural alterations in association tracts and language abilities in schoolchildren born very preterm and with poor fetal growth
Background: Prematurity and perinatal risk factors may influence white matter microstructure. In turn, these maturational changes may influence language development in this high-risk population of children. Objective: To evaluate differences in the microstructure of association tracts between preterm and term children and between preterm children with appropriate growth and those with fetal growth restriction and to study whether the diffusion tensor metrics of these tracts correlate with language abilities in schoolchildren with no severe neurological impairment. Materials and methods: This study prospectively followed 56 very preterm children (mean gestational age: 28.7 weeks) and 21 age- and gender-matched term children who underwent diffusion tensor imaging at a mean age of 9 years. We used automated probabilistic tractography and measured fractional anisotropy in seven bilateral association tracts known to belong to the white matter language network. Both groups participated in language assessment using five standardised tests at the same age. Results: Preterm children had lower fractional anisotropy in the right superior longitudinal fasciculus 1 compared to term children (P < 0.05). Preterm children with fetal growth restriction had lower fractional anisotropy in the left inferior longitudinal fasciculus compared to preterm children with appropriate fetal growth (P < 0.05). Fractional anisotropy in three dorsal tracts and in two dorsal and one ventral tract had a positive correlation with language assessments among preterm children and preterm children with fetal growth restriction, respectively (P < 0.05). Conclusion: There were some microstructural differences in language-related tracts between preterm and term children and between preterm children with appropriate and those with restricted fetal growth. Children with better language abilities had a higher fractional anisotropy in distinct white matter tracts.
- Open Access
- 1 June 2022
- Published by Medical Journals Sweden AB in Acta Orthopaedica
- Vol. 93, 519-527
Bioactive glass granules versus standard autologous and allogeneic bone grafts: a randomized trial of 49 adult bone tumor patients with a 10-year follow-up
Background and purpose: As a synthetic bone void filler, bioactive glasses (BGs) may enhance angiogenesis and osteogenesis. In this randomized trial, we compared the clinical efficacy of BG granules and standard bone grafts in patients undergoing surgery for benign bone tumors.Patients and methods: 49 recruited patients were randomized to receive BG granules or undergo conventional bone grafting to fill defects following tumor removal. As the standard of care, small-sized defects were filled with autologous graft, and large-sized defects were filled with allogeneic graft. The primary endpoint was treatment success at 1 year, defined by no reoperation, no tumor recurrence, and no device-related adverse events. Secondary endpoints included patient-reported outcomes (Rand-36 and pain scores) and quantitative assessment of blood flow and metabolic activity by means of 18F-fluoride PET/CT imaging. As an off-trial group, 15 children and adolescents (age < 18 years) underwent tumor removal and BG-filling, without randomization.Results: At 1-year, 21 of 25 BG-treated patients (risk estimate 0.84, 95% confidence interval [CI] 0.70–0.98) and 20 of 24 patients in the standard of care group (0.83, CI 0.68–0.98) met the criteria for treatment success. The groups had similar Rand-36 scores. In patients with small defects, BG filling was associated with shorter operative time and less postoperative pain at 1 month. In patients with large defects, blood flow was similar, but BG-filled defects showed higher metabolic activity than allograft-filled defects at 1-year. The survey of the postoperative period ≥10 years revealed no BG-related adverse events.Interpretation: BG granules had similar overall rates of treatment success compared with autografts and allografts, but large-scale trials are needed for the confirmation of clinical equivalence. The extended metabolic activity confirms the expected cellular responses of osseointegrated BG granules.
- 1 June 2022
- Published by Medical Journals Sweden AB in Acta Orthopaedica
- Vol. 93, 519-527
Open AccessBioactive glass granules versus standard autologous and allogeneic bone grafts: a randomized trial of 49 adult bone tumor patients with a 10-year follow-up
Background and purpose: As a synthetic bone void filler, bioactive glasses (BGs) may enhance angiogenesis and osteogenesis. In this randomized trial, we compared the clinical efficacy of BG granules and standard bone grafts in patients undergoing surgery for benign bone tumors.Patients and methods: 49 recruited patients were randomized to receive BG granules or undergo conventional bone grafting to fill defects following tumor removal. As the standard of care, small-sized defects were filled with autologous graft, and large-sized defects were filled with allogeneic graft. The primary endpoint was treatment success at 1 year, defined by no reoperation, no tumor recurrence, and no device-related adverse events. Secondary endpoints included patient-reported outcomes (Rand-36 and pain scores) and quantitative assessment of blood flow and metabolic activity by means of 18F-fluoride PET/CT imaging. As an off-trial group, 15 children and adolescents (age < 18 years) underwent tumor removal and BG-filling, without randomization.Results: At 1-year, 21 of 25 BG-treated patients (risk estimate 0.84, 95% confidence interval [CI] 0.70–0.98) and 20 of 24 patients in the standard of care group (0.83, CI 0.68–0.98) met the criteria for treatment success. The groups had similar Rand-36 scores. In patients with small defects, BG filling was associated with shorter operative time and less postoperative pain at 1 month. In patients with large defects, blood flow was similar, but BG-filled defects showed higher metabolic activity than allograft-filled defects at 1-year. The survey of the postoperative period ≥10 years revealed no BG-related adverse events.Interpretation: BG granules had similar overall rates of treatment success compared with autografts and allografts, but large-scale trials are needed for the confirmation of clinical equivalence. The extended metabolic activity confirms the expected cellular responses of osseointegrated BG granules.
- Open Access
- 24 April 2021
- Published by Springer Nature in Pediatric Radiology
- Vol. 51 (8) , 1471-1480
The impact of susceptibility correction on diffusion metrics in adolescents
Background: Diffusion tensor imaging is a widely used imaging method of brain white matter, but it is prone to imaging artifacts. The data corrections can affect the measured values.Objective: To explore the impact of susceptibility correction on diffusion metrics.Materials and methods: A cohort of 27 healthy adolescents (18 boys, 9 girls, mean age 12.7 years) underwent 3-T MRI, and we collected two diffusion data sets (anterior–posterior). The data were processed both with and without susceptibility artifact correction. We derived fractional anisotropy, mean diffusivity and histogram data of fiber length distribution from both the corrected and uncorrected data, which were collected from the corpus callosum, corticospinal tract and cingulum bilaterally.Results: Fractional anisotropy and mean diffusivity values significantly differed when comparing the pathways in all measured tracts. The fractional anisotropy values were lower and the mean diffusivity values higher in the susceptibility-corrected data than in the uncorrected data. We found a significant difference in total tract length in the corpus callosum and the corticospinal tract.Conclusion: This study indicates that susceptibility correction has a significant effect on measured fractional anisotropy, and on mean diffusivity values and tract lengths. To receive reliable and comparable results, the correction should be used systematically.
- 24 April 2021
- Published by Springer Nature in Pediatric Radiology
- Vol. 51 (8) , 1471-1480
Open AccessThe impact of susceptibility correction on diffusion metrics in adolescents
Background: Diffusion tensor imaging is a widely used imaging method of brain white matter, but it is prone to imaging artifacts. The data corrections can affect the measured values.Objective: To explore the impact of susceptibility correction on diffusion metrics.Materials and methods: A cohort of 27 healthy adolescents (18 boys, 9 girls, mean age 12.7 years) underwent 3-T MRI, and we collected two diffusion data sets (anterior–posterior). The data were processed both with and without susceptibility artifact correction. We derived fractional anisotropy, mean diffusivity and histogram data of fiber length distribution from both the corrected and uncorrected data, which were collected from the corpus callosum, corticospinal tract and cingulum bilaterally.Results: Fractional anisotropy and mean diffusivity values significantly differed when comparing the pathways in all measured tracts. The fractional anisotropy values were lower and the mean diffusivity values higher in the susceptibility-corrected data than in the uncorrected data. We found a significant difference in total tract length in the corpus callosum and the corticospinal tract.Conclusion: This study indicates that susceptibility correction has a significant effect on measured fractional anisotropy, and on mean diffusivity values and tract lengths. To receive reliable and comparable results, the correction should be used systematically.
- Open Access
- 20 August 2020
- Published by Springer Nature in Annals of Biomedical Engineering
- Vol. 49 (2) , 653-662
Quantification of Myocardial Blood Flow by Machine Learning Analysis of Modified Dual Bolus MRI Examination
Contrast-enhanced magnetic resonance imaging (MRI) is a promising method for estimating myocardial blood flow (MBF). However, it is often affected by noise from imaging artefacts, such as dark rim artefact obscuring relevant features. Machine learning enables extracting important features from such noisy data and is increasingly applied in areas where traditional approaches are limited. In this study, we investigate the capacity of machine learning, particularly support vector machines (SVM) and random forests (RF), for estimating MBF from tissue impulse response signal in an animal model. Domestic pigs (n = 5) were subjected to contrast enhanced first pass MRI (MRI-FP) and the impulse response at different regions of the myocardium (n = 24/pig) were evaluated at rest (n = 120) and stress (n = 96). Reference MBF was then measured using positron emission tomography (PET). Since the impulse response may include artefacts, classification models based on SVM and RF were developed to discriminate noisy signal. In addition, regression models based on SVM, RF and linear regression (for comparison) were developed for estimating MBF from the impulse response at rest and stress. The classification and regression models were trained on data from 4 pigs (n = 168) and tested on 1 pig (n = 48). Models based on SVM and RF outperformed linear regression, with higher correlation (R SVM 2 = 0.81, R RF 2 = 0.74, R linear_regression 2 = 0.60; ρSVM = 0.76, ρRF = 0.76, ρlinear_regression = 0.71) and lower error (RMSESVM = 0.67 mL/g/min, RMSERF = 0.77 mL/g/min, RMSElinear_regression = 0.96 mL/g/min) for predicting MBF from MRI impulse response signal. Classifier based on SVM was optimal for detecting impulse response signals with artefacts (accuracy = 92%). Modified dual bolus MRI signal, combined with machine learning, has potential for accurately estimating MBF at rest and stress states, even from signals with dark rim artefacts. This could provide a protocol for reliable and easy estimation of MBF, although further research is needed to clinically validate the approach.
- 20 August 2020
- Published by Springer Nature in Annals of Biomedical Engineering
- Vol. 49 (2) , 653-662
Open AccessQuantification of Myocardial Blood Flow by Machine Learning Analysis of Modified Dual Bolus MRI Examination
Contrast-enhanced magnetic resonance imaging (MRI) is a promising method for estimating myocardial blood flow (MBF). However, it is often affected by noise from imaging artefacts, such as dark rim artefact obscuring relevant features. Machine learning enables extracting important features from such noisy data and is increasingly applied in areas where traditional approaches are limited. In this study, we investigate the capacity of machine learning, particularly support vector machines (SVM) and random forests (RF), for estimating MBF from tissue impulse response signal in an animal model. Domestic pigs (n = 5) were subjected to contrast enhanced first pass MRI (MRI-FP) and the impulse response at different regions of the myocardium (n = 24/pig) were evaluated at rest (n = 120) and stress (n = 96). Reference MBF was then measured using positron emission tomography (PET). Since the impulse response may include artefacts, classification models based on SVM and RF were developed to discriminate noisy signal. In addition, regression models based on SVM, RF and linear regression (for comparison) were developed for estimating MBF from the impulse response at rest and stress. The classification and regression models were trained on data from 4 pigs (n = 168) and tested on 1 pig (n = 48). Models based on SVM and RF outperformed linear regression, with higher correlation (R SVM 2 = 0.81, R RF 2 = 0.74, R linear_regression 2 = 0.60; ρSVM = 0.76, ρRF = 0.76, ρlinear_regression = 0.71) and lower error (RMSESVM = 0.67 mL/g/min, RMSERF = 0.77 mL/g/min, RMSElinear_regression = 0.96 mL/g/min) for predicting MBF from MRI impulse response signal. Classifier based on SVM was optimal for detecting impulse response signals with artefacts (accuracy = 92%). Modified dual bolus MRI signal, combined with machine learning, has potential for accurately estimating MBF at rest and stress states, even from signals with dark rim artefacts. This could provide a protocol for reliable and easy estimation of MBF, although further research is needed to clinically validate the approach.
- Open Access
- 1 May 2020
- Published by Wolters Kluwer Health in Neurology Neuroimmunology & Neuroinflammation
- Vol. 7 (3)
Insights into disseminated MS brain pathology with multimodal diffusion tensor and PET imaging
Objective To evaluate in vivo the co-occurrence of microglial activation and microstructural white matter (WM) damage in the MS brain and to examine their association with clinical disability. Methods 18-kDa translocator protein (TSPO) brain PET imaging was performed for evaluation of microglial activation by using the radioligand [11C](R)-PK11195. TSPO binding was evaluated as the distribution volume ratio (DVR) from dynamic PET images. Diffusion tensor imaging (DTI) and conventional MRI (cMRI) were performed at the same time. Mean fractional anisotropy (FA) and mean (MD), axial, and radial (RD) diffusivities were calculated within the whole normal-appearing WM (NAWM) and segmented NAWM regions appearing normal in cMRI. Fifty-five patients with MS and 15 healthy controls (HCs) were examined. Results Microstructural damage was observed in the NAWM of the MS brain. DTI parameters of patients with MS were significantly altered in the NAWM compared with an age- and sex-matched HC group: mean FA was decreased, and MD and RD were increased. These structural abnormalities correlated with increased TSPO binding in the whole NAWM and in the temporal NAWM (p < 0.05 for all correlations; p < 0.01 for RD in the temporal NAWM). Both compromised WM integrity and increased microglial activation in the NAWM correlated significantly with higher clinical disability measured with the Expanded Disability Status Scale score. Conclusions Widespread structural disruption in the NAWM is linked to neuroinflammation, and both phenomena associate with clinical disability. Multimodal PET and DTI allow in vivo evaluation of widespread MS pathology not visible using cMRI.
- 1 May 2020
- Published by Wolters Kluwer Health in Neurology Neuroimmunology & Neuroinflammation
- Vol. 7 (3)
Open AccessInsights into disseminated MS brain pathology with multimodal diffusion tensor and PET imaging
Objective To evaluate in vivo the co-occurrence of microglial activation and microstructural white matter (WM) damage in the MS brain and to examine their association with clinical disability. Methods 18-kDa translocator protein (TSPO) brain PET imaging was performed for evaluation of microglial activation by using the radioligand [11C](R)-PK11195. TSPO binding was evaluated as the distribution volume ratio (DVR) from dynamic PET images. Diffusion tensor imaging (DTI) and conventional MRI (cMRI) were performed at the same time. Mean fractional anisotropy (FA) and mean (MD), axial, and radial (RD) diffusivities were calculated within the whole normal-appearing WM (NAWM) and segmented NAWM regions appearing normal in cMRI. Fifty-five patients with MS and 15 healthy controls (HCs) were examined. Results Microstructural damage was observed in the NAWM of the MS brain. DTI parameters of patients with MS were significantly altered in the NAWM compared with an age- and sex-matched HC group: mean FA was decreased, and MD and RD were increased. These structural abnormalities correlated with increased TSPO binding in the whole NAWM and in the temporal NAWM (p < 0.05 for all correlations; p < 0.01 for RD in the temporal NAWM). Both compromised WM integrity and increased microglial activation in the NAWM correlated significantly with higher clinical disability measured with the Expanded Disability Status Scale score. Conclusions Widespread structural disruption in the NAWM is linked to neuroinflammation, and both phenomena associate with clinical disability. Multimodal PET and DTI allow in vivo evaluation of widespread MS pathology not visible using cMRI.