Fang-Fang Yin

Overview:

Stereotactic radiosurgery, Stereotactic body radiation therapy, treatment planning optimization, knowledge guided radiation therapy, intensity-modulated radiation therapy, image-guided radiation therapy, oncological imaging and informatics

Positions:

Professor in Radiation Oncology

Radiation Oncology
School of Medicine

Professor of Medical Physics at Duke Kunshan University

DKU Faculty
Duke Kunshan University

Member of the Duke Cancer Institute

Duke Cancer Institute
School of Medicine

Education:

B.S. 1982

Zhejiang University (China)

M.S. 1987

Bowling Green State University

Ph.D. 1992

University of Chicago

Certificate In Therapeutic Radiologic Physics, Radiation Physics

American Board of Radiology

Grants:

Motion Management Using 4D-MRI for Liver Cancer in Radiation Therapy

Administered By
Radiation Oncology
Awarded By
National Institutes of Health
Role
Co-Principal Investigator
Start Date
End Date

Digital tomosynthesis: a new paradigm for radiation treatment verification

Administered By
Radiation Oncology
Awarded By
National Institutes of Health
Role
Principal Investigator
Start Date
End Date

Robotic SPECT for Biological Imaging Onboard Radiation Therapy Machines

Administered By
Radiation Oncology
Awarded By
National Institutes of Health
Role
Co-Principal Investigator
Start Date
End Date

Accurate, High Resolution 3D Dosimetry

Administered By
Radiation Oncology
Awarded By
National Institutes of Health
Role
Collaborator
Start Date
End Date

A Limited-angle Intra-fractional Verification (LIVE) System for SBRT Treatments

Administered By
Radiation Oncology
Awarded By
National Institutes of Health
Role
Co-Principal Investigator
Start Date
End Date

Publications:

Development of a computerized portal verification system for radiation therapy of breast cancers

© 1995 SPIE. All rights reserved. A fully automated system is being developed for the portal verification of tangential breast fields in radiation therapy of breast cancer. The automated verification system involves image acquisition, image feature extraction, feature correlation between reference and portal images, and quantitative evaluation of patient setup. In this study, the portal images are acquired using a matrix liquid ion-chamber electronic portal imaging device (EPID), and have a matrix size of 256 × 256 pixels with 12-bit gray levels. A hierarchical region processing technique is developed to extract poor contrast features in the portal image generated by megavoltage photon beams at different levels sequentially. The treatment field is initially extracted from the portal image. The skin line is then extracted from the treatment field. Finally, the lung/soft tissue separation is extracted from the breast region. A Chamfer matching filter is used to correlate features in the portal image with those in the reference image. The resulting parameters for rotation, translation and scaling are used for the setup evaluation of the treatment field.
Authors
Yin, FF; Lai, W; Chen, CW; Nelson, DF; Schell, MC; Rubin, P
MLA Citation
Yin, F. F., et al. “Development of a computerized portal verification system for radiation therapy of breast cancers.” Proceedings of Spie  the International Society for Optical Engineering, vol. 2434, 1995, pp. 540–45. Scopus, doi:10.1117/12.208725.
URI
https://scholars.duke.edu/individual/pub1425334
Source
scopus
Published In
Smart Structures and Materials 2005: Active Materials: Behavior and Mechanics
Volume
2434
Published Date
Start Page
540
End Page
545
DOI
10.1117/12.208725

An investigation of machine learning methods in delta-radiomics feature analysis.

PURPOSE: This study aimed to investigate the effectiveness of using delta-radiomics to predict overall survival (OS) for patients with recurrent malignant gliomas treated by concurrent stereotactic radiosurgery and bevacizumab, and to investigate the effectiveness of machine learning methods for delta-radiomics feature selection and building classification models. METHODS: The pre-treatment, one-week post-treatment, and two-month post-treatment T1 and T2 fluid-attenuated inversion recovery (FLAIR) MRI were acquired. 61 radiomic features (intensity histogram-based, morphological, and texture features) were extracted from the gross tumor volume in each image. Delta-radiomics were calculated between the pre-treatment and post-treatment features. Univariate Cox regression and 3 multivariate machine learning methods (L1-regularized logistic regression [L1-LR], random forest [RF] or neural networks [NN]) were used to select a reduced number of features, and 7 machine learning methods (L1-LR, L2-LR, RF, NN, kernel support vector machine [KSVM], linear support vector machine [LSVM], or naïve bayes [NB]) was used to build classification models for predicting OS. The performances of the total 21 model combinations built based on single-time-point radiomics (pre-treatment, one-week post-treatment, and two-month post-treatment) and delta-radiomics were evaluated by the area under the receiver operating characteristic curve (AUC). RESULTS: For a small cohort of 12 patients, delta-radiomics resulted in significantly higher AUC than pre-treatment radiomics (p-value<0.01). One-week/two-month delta-features resulted in significantly higher AUC (p-value<0.01) than the one-week/two-month post-treatment features, respectively. 18/21 model combinations were with higher AUC from one-week delta-features than two-month delta-features. With one-week delta-features, RF feature selector + KSVM classifier and RF feature selector + NN classifier showed the highest AUC of 0.889. CONCLUSIONS: The results indicated that delta-features could potentially provide better treatment assessment than single-time-point features. The treatment assessment is substantially affected by the time point for computing the delta-features and the combination of machine learning methods for feature selection and classification.
Authors
Chang, Y; Lafata, K; Sun, W; Wang, C; Chang, Z; Kirkpatrick, JP; Yin, F-F
MLA Citation
Chang, Yushi, et al. “An investigation of machine learning methods in delta-radiomics feature analysis..” Plos One, vol. 14, no. 12, 2019. Pubmed, doi:10.1371/journal.pone.0226348.
URI
https://scholars.duke.edu/individual/pub1423241
PMID
31834910
Source
pubmed
Published In
Plos One
Volume
14
Published Date
Start Page
e0226348
DOI
10.1371/journal.pone.0226348

Technical Note: Investigation of the dosimetric impact of stray radiation on the Common Control Unit of the IBA Blue Phantom2.

PURPOSE: This technical note aims to investigate the dosimetric impact of stray radiation on the Common Control Unit (CCU) of the IBA Blue Phantom2 and the measured beam data. METHODS: Three CCUs of the same model were used for the study. The primary test CCU was placed at five distances from the radiation beam central axis. At each distance, a set of depth dose and beam profiles for two open and two wedge fields were measured. The field sizes were 10 × 10 cm2 and 30 × 30 cm2 for the open fields, and 30 × 30 cm2 and 15 × 15 cm2 for the 30° and 60° wedges, respectively. The other two CCUs were used to cross check the data of the primary CCU. Assuming the effect of stray radiation on the data measured at the farthest reachable distance 4.5 m is negligible, the dosimetric impact of stray radiation on the CCU and consequently on the measured data can be extracted for analysis by comparing it with those measured at shorter distances. RESULTS: The results of three CCUs were consistent. The dosimetric impact of stray radiation was greater for lower energies at larger field sizes. For open fields, the data variation was up to 4.5% for depth dose curves and 7.1% for beam profiles. For wedge fields, the data variation was up to 9.3% for depth dose curves and 10.6% for beam profiles. Moreover, for wedge field profiles in the wedge direction, they became flatter as the CCU was placed closer to the primary radiation beam, manifesting smaller wedge angles. CONCLUSION: The stray radiation added a uniform background noise on all measured data. The magnitude of the noise is inversely proportional to the square of the distance of the CCU to the primary radiation beam, approximately following the inverse square law.
Authors
Cui, G; Duan, J; Yang, Y; Yin, F-F
MLA Citation
Cui, Guoqiang, et al. “Technical Note: Investigation of the dosimetric impact of stray radiation on the Common Control Unit of the IBA Blue Phantom2..” J Appl Clin Med Phys, Nov. 2019. Pubmed, doi:10.1002/acm2.12769.
URI
https://scholars.duke.edu/individual/pub1421792
PMID
31729812
Source
pubmed
Published In
Journal of Applied Clinical Medical Physics
Published Date
DOI
10.1002/acm2.12769

Probability-based 3D k-space sorting for motion robust 4D-MRI.

Background: Current 4D-MRI techniques are prone to breathing-variation-induced motion artifacts. This study developed a novel method for motion-robust multi-cycle 4D-MRI using probability-based multi-cycle sorting to overcome this deficiency. Methods: The main cycles were first extracted from the breathing signal. 3D k-space data were then sorted using a result-driven method for each main cycle. The new method was tested on a 4D-extended cardiac-torso (XCAT) phantom with a patient and an artificially generated breathing curve. For comparison, the k-space data were sorted using conventional phase sorting to generate single-cycle 4D-MRI images. Signal-to-noise ratio (SNR) of tumor and liver, tumor volume consistency, and average intensity projection (AIP) accuracy were compared between the two methods. The original phantom images were used as references for the evaluation. Results: The new method showed improved tumor-to-liver SNR and tumor volume consistency as compared to 3D k-space phase sorting in both the simulated artificial and real patient breathing signals. For the artificial breathing cycles, the average tumor-to-liver SNR and standard deviation (SD) of tumor volume were 2.53 and 3.80% for cycle 1, 2.24 and 6.16% for cycle 2 of probability-based sorting as compared to 1.47 and 21.83% obtained using the phase sorting method; for the patient breathing curve, values of 1.99 and 2.71%, 1.97 and 3.29%, 1.88 and 4.16% were observed for cycle 1, cycle 2 and cycle 3 of probability-based sorting, versus 1.44 and 7.20% for phase sorting method. Furthermore, the AIP accuracy was improved in the probability-based sorting approach when compared to phase sorting, with the average intensity difference per voxel reduced from 0.39 to 0.15 for the artificial curve, and from 0.46 to 0.21 for the patient curve. Conclusions: We demonstrated the feasibility of probability-based 3D k-space sorting for motion-robust multi-cycle 4D-MRI reconstruction with breathing variation induced motion artifact reduction compared with conventional 2D image sorting and 3D phase sorting methods. This new technique can potentially improve the accuracy of radiation treatment guidance for mobile targets.
Authors
Sun, D; Liang, X; Yin, F; Cai, J
MLA Citation
Sun, Duohua, et al. “Probability-based 3D k-space sorting for motion robust 4D-MRI..” Quant Imaging Med Surg, vol. 9, no. 7, July 2019, pp. 1326–36. Pubmed, doi:10.21037/qims.2019.07.06.
URI
https://scholars.duke.edu/individual/pub1404444
PMID
31448217
Source
pubmed
Published In
Quantitative Imaging in Medicine and Surgery
Volume
9
Published Date
Start Page
1326
End Page
1336
DOI
10.21037/qims.2019.07.06

Feasibility of radiosurgery dosimetry using NIPAM 3D dosimeters and x-ray CT

© Published under licence by IOP Publishing Ltd. We investigated the feasibility of using N-isopropylacrylamide (NIPAM) dosimeters with x-ray CT to verify radiosurgery dose. Dosimeters were prepared at one facility and shipped to a second facility for irradiation. A simulation CT was acquired and plans prepared for a 4 field box, and a 4 arc VMAT radiosurgery plan to 6 targets with 1cm diameter. Each dosimeter was aligned via CBCT and irradiated, followed by 5 diagnostic CTs acquired after >24 hours, which were averaged for analysis. Absolute dose calibration was applied and dose evaluated for both plans. Hounsfield Units were proportional to dose above 10-12Gy. For the 4-field box, mean difference between measured and predicted dose >10Gy was -0.13Gy -1.69Gy and gamma index was <1 for 72% and 65% of voxels using a 5% / 1mm and 3% / 2mm criteria, respectively (threshold = 15Gy, global dose criteria). For the multifocal SRS case, mean dose within each target was within -0.14Gy- 0.55Gy of the expected value, and gamma index was < 1 for 94.0% and 99.5% of voxels, respectively (threshold = 15Gy). NIPAM based 3D dosimetry with x-ray CT is well suited for validating radiosurgery spatial alignment, as well as dose distributions when dose is above 10-12Gy.
Authors
Adamson, J; Carroll, J; Trager, M; Yoon, P; Kodra, J; Yin, FF; Maynard, E; Hilts, M; Oldham, M; Jirasik, A
MLA Citation
Adamson, J., et al. “Feasibility of radiosurgery dosimetry using NIPAM 3D dosimeters and x-ray CT.” Journal of Physics: Conference Series, vol. 1305, no. 1, 2019. Scopus, doi:10.1088/1742-6596/1305/1/012004.
URI
https://scholars.duke.edu/individual/pub1417159
Source
scopus
Published In
Journal of Physics: Conference Series
Volume
1305
Published Date
DOI
10.1088/1742-6596/1305/1/012004

Research Areas:

Bioinformatics
Medical physics