Artificial intelligence for detecting small FDG-positive lung nodules in digital PET/CT: impact of image reconstructions on diagnostic performance

Abstract Background: The superior accuracy and sensitivity of 18F-FDG-PET/CT in comparison to morphological imaging alone leads to an upstaging in up to 30 % of lymphoma patients. Novel digital PET/CT scanners might enable to reduce administered tracer activity or scan time duration while maintaining diagnostic performance; this might allow for a higher patient throughput or a reduced radiation exposure, respectively. In particular, the radiation exposure reduction is of interest due to the often young age and high remission rate of lymphoma patients.Methods: Twenty patients with (suspected) lymphoma (6 for initial staging, 12 after systemic treatment, 2 in suspicion of recurrence) sequentially underwent 18F-FDG-PET/CT examinations on a digital PET/CT (Siemens Biograph Vision) with a total scan time duration of 15 minutes (reference acquisition protocol) and 5 minutes (reduced acquisition protocol) using continuous-bed-motion. Both data sets were reconstructed using either standalone time of flight (TOF) or in combination with point spread function (PSF), each with 2 and 4 iterations. Lesion detectability by blinded assessment (separately for supra- and infradiaphragmal nodal lesions and for extranodal lesions), lesion image quantification, and image noise were used as metrics to assess diagnostic performance. Additionally, Deauville Score was compared for all patients after systemic treatment.Results: All defined regions were correctly classified in the images acquired with reduced emission time, and therefore, no changes in staging were observed. Lesion quantification was acceptable, that is, mean absolute percentage deviation of maximum and peak standardized uptake values were 6.8% and 6.4% (derived from 30 lesions). A threefold reduction of scan time duration led to an increase in image noise from 7.1% to 11.0% (images reconstructed with 4 iterations) and from 4.7% to 7.2% (images reconstructed with 2 iterations). No deviations in Deauville Score were observed.Conclusion: These results suggest that scan time duration or administered tracer activity can be reduced threefold without compromising diagnostic performance. Especially a reduction of administered activity might allow for a lower radiation exposure and better health economics. Larger trials are warranted to confirm our results.

Download Full-text

Evaluation of 18F-FDG PET/CT images acquired with a reduced scan time duration in lymphoma patients using the digital biograph vision

BMC Cancer ◽

10.1186/s12885-020-07723-2 ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Manuel Weber ◽

Walter Jentzen ◽

Regina Hofferber ◽

Ken Herrmann ◽

Wolfgang Peter Fendler ◽

...

Keyword(s):

Radiation Exposure ◽

Fdg Pet ◽

Diagnostic Performance ◽

Systemic Treatment ◽

Time Duration ◽

Tracer Activity ◽

Scan Time ◽

Pet Ct ◽

Digital Pet ◽

Fdg Pet Ct

Abstract Background The superior accuracy and sensitivity of 18F-FDG-PET/CT in comparison to morphological imaging alone leads to an upstaging in up to 30% of lymphoma patients. Novel digital PET/CT scanners might enable to reduce administered tracer activity or scan time duration while maintaining diagnostic performance; this might allow for a higher patient throughput or a reduced radiation exposure, respectively. In particular, the radiation exposure reduction is of interest due to the often young age and high remission rate of lymphoma patients. Methods Twenty patients with (suspected) lymphoma (6 for initial staging, 12 after systemic treatment, 2 in suspicion of recurrence) sequentially underwent 18F-FDG-PET/CT examinations on a digital PET/CT (Siemens Biograph Vision) with a total scan time duration of 15 min (reference acquisition protocol) and 5 min (reduced acquisition protocol) using continuous-bed-motion. Both data sets were reconstructed using either standalone time of flight (TOF) or in combination with point spread function (PSF), each with 2 and 4 iterations. Lesion detectability by blinded assessment (separately for supra- and infradiaphragmal nodal lesions and for extranodal lesions), lesion image quantification, and image noise were used as metrics to assess diagnostic performance. Additionally, Deauville Score was compared for all patients after systemic treatment. Results All defined regions were correctly classified in the images acquired with reduced emission time, and therefore, no changes in staging were observed. Lesion quantification was acceptable, that is, mean absolute percentage deviation of maximum and peak standardized uptake values were 6.8 and 6.4% (derived from 30 lesions). A threefold reduction of scan time duration led to an increase in image noise from 7.1 to 11.0% (images reconstructed with 4 iterations) and from 4.7 to 7.2% (images reconstructed with 2 iterations). No deviations in Deauville Score were observed. Conclusion These results suggest that scan time duration or administered tracer activity can be reduced threefold without compromising diagnostic performance. Especially a reduction of administered activity might allow for a lower radiation exposure and better health economics. Larger trials are warranted to confirm our results.

Download Full-text

Evaluation of 18F-FDG PET/CT Images Acquired with a Reduced Scan Time Duration in Lymphoma Patients using the Digital Biograph Vision

10.21203/rs.3.rs-46922/v2 ◽

2020 ◽

Author(s):

Manuel Weber ◽

Walter Jentzen ◽

Regina Hofferber ◽

Ken Herrmann ◽

Wolfgang Peter Fendler ◽

...

Keyword(s):

Radiation Exposure ◽

Fdg Pet ◽

Diagnostic Performance ◽

Systemic Treatment ◽

Time Duration ◽

Tracer Activity ◽

Scan Time ◽

Pet Ct ◽

Digital Pet ◽

Fdg Pet Ct

Abstract Background The superior accuracy and sensitivity of 18 F-FDG-PET/CT in comparison to morphological imaging alone leads to an upstaging in up to 30% of lymphoma patients. Novel digital PET/CT scanners might enable to reduce administered tracer activity or scan time duration while maintaining diagnostic performance; this might allow for a higher patient throughput or a reduced radiation exposure, respectively. In particular, the radiation exposure reduction is of interest due to the often young age and high remission rate of lymphoma patients. Methods Twenty patients with (suspected) lymphoma (6 for initial staging, 12 after systemic treatment, 2 in suspicion of recurrence) sequentially underwent 18 F-FDG-PET/CT examinations on a digital PET/CT (Siemens Biograph Vision) with a total scan time duration of 15 minutes (reference acquisition protocol) and 5 minutes (reduced acquisition protocol) using continuous-bed-motion. Both data sets were reconstructed using either standalone time of flight (TOF) or in combination with point spread function (PSF), each with 2 and 4 iterations. Lesion detectability by blinded assessment (separately for supra- and infradiaphragmal nodal lesions and for extranodal lesions), lesion image quantification, and image noise were used as metrics to assess diagnostic performance. Additionally, Deauville Score was compared for all patients after systemic treatment. Results All defined regions were correctly classified in the images acquired with reduced emission time, and therefore, no changes in staging were observed. Lesion quantification was acceptable, that is, mean absolute percentage deviation of maximum and peak standardized uptake values were 6.8% and 6.4% (derived from 30 lesions). A threefold reduction of scan time duration led to an increase in image noise from 7.1% to 11.0% (images reconstructed with 4 iterations) and from 4.7% to 7.2% (images reconstructed with 2 iterations). No deviations in Deauville Score were observed. Conclusion These results suggest that scan time duration or administered tracer activity can be reduced threefold without compromising diagnostic performance. Especially a reduction of administered activity might allow for a lower radiation exposure and better health economics. Larger trials are warranted to confirm our results.

Download Full-text

Evaluation of 18F-FDG PET/CT Images Acquired with a Reduced Scan Time Duration in Lymphoma Patients using the Digital Biograph Vision

10.21203/rs.3.rs-46922/v3 ◽

2021 ◽

Author(s):

Manuel Weber ◽

Walter Jentzen ◽

Regina Hofferber ◽

Ken Herrmann ◽

Wolfgang Peter Fendler ◽

...

Keyword(s):

Radiation Exposure ◽

Fdg Pet ◽

Diagnostic Performance ◽

Systemic Treatment ◽

Time Duration ◽

Tracer Activity ◽

Scan Time ◽

Pet Ct ◽

Digital Pet ◽

Fdg Pet Ct

Abstract Background The superior accuracy and sensitivity of 18 F-FDG-PET/CT in comparison to morphological imaging alone leads to an upstaging in up to 30% of lymphoma patients. Novel digital PET/CT scanners might enable to reduce administered tracer activity or scan time duration while maintaining diagnostic performance; this might allow for a higher patient throughput or a reduced radiation exposure, respectively. In particular, the radiation exposure reduction is of interest due to the often young age and high remission rate of lymphoma patients. Methods Twenty patients with (suspected) lymphoma (6 for initial staging, 12 after systemic treatment, 2 in suspicion of recurrence) sequentially underwent 18 F-FDG-PET/CT examinations on a digital PET/CT (Siemens Biograph Vision) with a total scan time duration of 15 minutes (reference acquisition protocol) and 5 minutes (reduced acquisition protocol) using continuous-bed-motion. Both data sets were reconstructed using either standalone time of flight (TOF) or in combination with point spread function (PSF), each with 2 and 4 iterations. Lesion detectability by blinded assessment (separately for supra- and infradiaphragmal nodal lesions and for extranodal lesions), lesion image quantification, and image noise were used as metrics to assess diagnostic performance. Additionally, Deauville Score was compared for all patients after systemic treatment. Results All defined regions were correctly classified in the images acquired with reduced emission time, and therefore, no changes in staging were observed. Lesion quantification was acceptable, that is, mean absolute percentage deviation of maximum and peak standardized uptake values were 6.8% and 6.4% (derived from 30 lesions). A threefold reduction of scan time duration led to an increase in image noise from 7.1% to 11.0% (images reconstructed with 4 iterations) and from 4.7% to 7.2% (images reconstructed with 2 iterations). No deviations in Deauville Score were observed. Conclusion These results suggest that scan time duration or administered tracer activity can be reduced threefold without compromising diagnostic performance. Especially a reduction of administered activity might allow for a lower radiation exposure and better health economics. Larger trials are warranted to confirm our results.

Download Full-text

Evaluation of 18F-FDG PET/CT Images Acquired with a Reduced Scan Time Duration in Lymphoma Patients using the Digital Biograph Vision

10.21203/rs.3.rs-46922/v1 ◽

2020 ◽

Author(s):

Manuel Weber ◽

Walter Jentzen ◽

Regina Hofferber ◽

Ken Herrmann ◽

Wolfgang Peter Fendler ◽

...

Keyword(s):

Radiation Exposure ◽

Fdg Pet ◽

Diagnostic Performance ◽

Systemic Treatment ◽

Time Duration ◽

Tracer Activity ◽

Scan Time ◽

Pet Ct ◽

Digital Pet ◽

Fdg Pet Ct

Abstract Background: The superior accuracy and sensitivity of 18F-FDG-PET/CT in comparison to morphological imaging alone leads to an upstaging in up to 30 % of lymphoma patients. Novel digital PET/CT scanners might enable to reduce administered tracer activity or scan time duration while maintaining diagnostic performance; this might allow for a higher patient throughput or a reduced radiation exposure, respectively. In particular, the radiation exposure reduction is of interest due to the often young age and high remission rate of lymphoma patients.Methods: Twenty patients with (suspected) lymphoma (6 for initial staging, 12 after systemic treatment, 2 in suspicion of recurrence) sequentially underwent 18F-FDG-PET/CT examinations on a digital PET/CT (Siemens Biograph Vision) with a total scan time duration of 15 minutes (reference acquisition protocol) and 5 minutes (reduced acquisition protocol) using continuous-bed-motion. Both data sets were reconstructed using either standalone time of flight (TOF) or in combination with point spread function (PSF), each with 2 and 4 iterations. Lesion detectability by blinded assessment (separately for supra- and infradiaphragmal nodal lesions and for extranodal lesions), lesion image quantification, and image noise were used as metrics to assess diagnostic performance. Additionally, Deauville Score was compared for all patients after systemic treatment.Results: All defined regions were correctly classified in the images acquired with reduced emission time, and therefore, no changes in staging were observed. Lesion quantification was acceptable, that is, mean absolute percentage deviation of maximum and peak standardized uptake values were 6.8% and 6.4% (derived from 30 lesions). A threefold reduction of scan time duration led to an increase in image noise from 7.1% to 11.0% (images reconstructed with 4 iterations) and from 4.7% to 7.2% (images reconstructed with 2 iterations). No deviations in Deauville Score were observed.Conclusion: These results suggest that scan time duration or administered tracer activity can be reduced threefold without compromising diagnostic performance. Especially a reduction of administered activity might allow for a lower radiation exposure and better health economics. Larger trials are warranted to confirm our results.

Download Full-text

Emission time optimisation for Ga-68-PSMA with a digital PET/CT – a phantom study

10.1055/s-0040-1708281 ◽

2020 ◽

Author(s):

PF Costa ◽

F Süßelbeck ◽

A Bramer ◽

M Conti ◽

M Weber ◽

...

Keyword(s):

Phantom Study ◽

Pet Ct ◽

Emission Time ◽

Digital Pet

Download Full-text

Diagnostic performance of 18F-FDG PET/CT, ultrasonography and MRI

Nuklearmedizin ◽

10.3413/nukmed-0605-13-06 ◽

2014 ◽

Vol 53 (03) ◽

pp. 89-94 ◽

Cited By ~ 17

Author(s):

D. H. Lee ◽

J.-K Yoon ◽

S. J. Lee ◽

T. H. Kim ◽

D. K. Kang ◽

...

Keyword(s):

Lymph Node ◽

Fdg Pet ◽

Axillary Lymph Node ◽

Diagnostic Performance ◽

Axillary Lymph ◽

Diagnostic Ability ◽

Node Metastasis ◽

Pet Ct ◽

18F Fdg ◽

Fdg Pet Ct

SummaryThe aim of this study was to evaluate the diagnostic abilities of 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) compared with those of ultrasonography and magnetic resonance imaging (MRI) for axillary lymph node staging in breast cancer patients. Patients, methods: Pre- operative 18F-FDG PET/non-contrast CT, ultrasonography and MRI were performed in 215 women with breast cancer. Axillary lymph node dissection was performed in all patients and the diagnostic performance of each modality was evaluated using histopathologic assessments as the reference standard. ROC curves were compared to evaluate the diagnostic ability of several imaging modalities (i. e., ultrasonography, MRI and 18F-FDG PET/CT). Results: In total, 132 patients (61.4%) had axillary lymph node metastasis. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for the detection of axillary lymph node metastasis were 72.3%, 77.3%, 66.7%, 81.6%, 75.3% for ultrasonography, 67.5%, 78.0%, 65.9%, 79.2%, 74.0% for MRI, and 62.7%, 88.6%, 77.6%, 79.1%, 78.6% for 18F-FDG PET/CT, respectively. There was no significant difference in diagnostic ability among the imaging modalities (i.e., ultrasonography, MRI and 18F-FDG PET/CT). The diagnostic ability of 18F-FDG PET/CT was significantly improved by combination with MRI (p = 0.0002) or ultrasonography (p < 0.0001). The combination of 18F-FDG PET/CT with ultrasonography had a similar diagnostic ability to that of all three modalities combined (18F-FDG PET/CT+ultraso- nography+MRI, p = 0.05). Conclusion: The diagnostic performance of 18F-FDG PET/CT for detection of axillary node metastasis was not significantly different from that of ultrasonography or MRI in breast cancer patients. Combining 18F-FDG PET/CT with ultrasonography or MRI could improve the diagnostic performance compared to 18F-FDG PET/CT alone.

Download Full-text

Artificial intelligence could alert for focal skeleton/bone marrow uptake in Hodgkin’s lymphoma patients staged with FDG-PET/CT

Scientific Reports ◽

10.1038/s41598-021-89656-9 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

May Sadik ◽

Jesús López-Urdaneta ◽

Johannes Ulén ◽

Olof Enqvist ◽

Armin Krupic ◽

...

Keyword(s):

Artificial Intelligence ◽

Bone Marrow ◽

Fdg Pet ◽

Test Group ◽

Observer Agreement ◽

Hodgkin's Lymphoma ◽

Bone Marrow Uptake ◽

Focal Uptake ◽

Pet Ct ◽

Fdg Pet Ct

AbstractTo develop an artificial intelligence (AI)-based method for the detection of focal skeleton/bone marrow uptake (BMU) in patients with Hodgkin’s lymphoma (HL) undergoing staging with FDG-PET/CT. The results of the AI in a separate test group were compared to the interpretations of independent physicians. The skeleton and bone marrow were segmented using a convolutional neural network. The training of AI was based on 153 un-treated patients. Bone uptake significantly higher than the mean BMU was marked as abnormal, and an index, based on the total squared abnormal uptake, was computed to identify the focal uptake. Patients with an index above a predefined threshold were interpreted as having focal uptake. As the test group, 48 un-treated patients who had undergone a staging FDG-PET/CT between 2017–2018 with biopsy-proven HL were retrospectively included. Ten physicians classified the 48 cases regarding focal skeleton/BMU. The majority of the physicians agreed with the AI in 39/48 cases (81%) regarding focal skeleton/bone marrow involvement. Inter-observer agreement between the physicians was moderate, Kappa 0.51 (range 0.25–0.80). An AI-based method can be developed to highlight suspicious focal skeleton/BMU in HL patients staged with FDG-PET/CT. Inter-observer agreement regarding focal BMU is moderate among nuclear medicine physicians.

Download Full-text