Jihye Baek

Jihye Baek

Auto-generated profile, based on publicly available information

Jihye Baek

Department of Radiology, Stanford University School of Medicine, Stanford, CA, USA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 January 2022 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Verified
Journal Article
Closed access
JBJihye Baek
Jihye Baek
LBLokesh Basavarajappa
Lokesh Basavarajappa
KHKenneth Hoyt
Kenneth Hoyt
KPKevin Parker
Kevin Parker
In medical imaging, quantitative measurements have shown promise in identifying diseases by classifying normal versus pathological parameters from tissues. The support vector machine (SVM) has shown promise as a supervised classification algorithm and has been widely used. However, the classification results typically identify a category of abnormal tissues but do not necessarily differentiate progressive stages of a disease. Moreover, the classification result is typically provided independently as a supplement to medical images, which contributes to an overload of information sources in the clinic. Hence, we propose a new imaging method utilizing the SVM to integrate classification results into medical images. This framework is called disease-specific imaging (DSI) that produces a color overlaid highlight on B-mode ultrasound images indicating the type, location, and severity of pathology from different conditions. In this article, the SVM training was performed to construct hyperplanes that can differentiate normal, fibrosis, steatosis, and pancreatic ductal adenocarcinoma (PDAC) metastases in livers based on ultrasound echoes. Also, cluster centroids for specific diseases define unique disease axes, and the inner product between measured features and any disease axis selected by the SVM quantifies the disease progression. The features were measured from 2794 ultrasound frames using the H-scan analysis, attenuation estimation, and B-mode image analysis. The performance of our proposed DSI method was evaluated for a preclinical model of steatosis ( ${n} =400$ frames). The contribution of each feature was assessed, and the results were compared with ground truth from histology. Moreover, the images generated by our DSI were compared with earlier imaging methods of B-mode, H-scan, and histology. The comparisons demonstrate that DSI images yield higher sensitivity to monitor progressive steatosis than B-mode and H-scan and provide a comparable performance with the histology. For the parameter comparison, DSI and H-scan resulted in similar correlation with histology ( ${r}_{s} =0.83$ ) but higher than attenuation ( ${r}_{s} =0.73$ ) and B-mode ( ${r}_{s} =0.47$ ). Therefore, we conclude that DSI utilizing the SVM applied to steatosis can visually represent the classification results with color highlighting, which can simplify the interpretation of classification compared to the traditional SVM result. We expect that the proposed DSI can be used for any medical imaging modality that can estimate multiple quantitative parameters at high resolution.

7 November 2022 in Machine Learning: Science and Technology
Verified
Journal Article
Open access
JBJihye Baek
Jihye Baek
AOAvice M O’cOnnell
Avice M O’cOnnell
KPKevin Parker
Kevin Parker
The improved diagnostic accuracy of ultrasound breast examinations remains an important goal. In this study, we propose a biophysical feature-based machine learning method for breast cancer detection to improve the performance beyond a benchmark deep learning algorithm and to furthermore provide a color overlay visual map of the probability of malignancy within a lesion. This overall framework is termed disease-specific imaging. Previously, 150 breast lesions were segmented and classified utilizing a modified fully convolutional network and a modified GoogLeNet, respectively. In this study multiparametric analysis was performed within the contoured lesions. Features were extracted from ultrasound radiofrequency, envelope, and log-compressed data based on biophysical and morphological models. The support vector machine with a Gaussian kernel constructed a nonlinear hyperplane, and we calculated the distance between the hyperplane and each feature’s data point in multiparametric space. The distance can quantitatively assess a lesion and suggest the probability of malignancy that is color-coded and overlaid onto B-mode images. Training and evaluation were performed on in vivo patient data. The overall accuracy for the most common types and sizes of breast lesions in our study exceeded 98.0% for classification and 0.98 for an area under the receiver operating characteristic curve, which is more precise than the performance of radiologists and a deep learning system. Further, the correlation between the probability and Breast Imaging Reporting and Data System enables a quantitative guideline to predict breast cancer. Therefore, we anticipate that the proposed framework can help radiologists achieve more accurate and convenient breast cancer classification and detection.

11 September 2021
Verified
Conference Paper
Closed access
JBJihye Baek
Jihye Baek
SQShuyang Qin
Shuyang Qin
PPPeter A. Prieto
Peter A. Prieto
KPKevin Parker
Kevin Parker
The H-scan analysis has been developed as a sensitive measure of ultrasound scattering types. In this study, we aim to distinguish between two different melanoma tumors: YUMM and YUMMER. To characterize the tumor tissues, H-scan was applied to the melanoma samples, resulting in estimation of several parameters. Multiparametric analysis was performed by measuring the H-scan blue fraction, attenuation coefficient, H-scan signal-to-noise ratio (SNR), and B-scan SNR. Principal component analysis was applied to the parameters, forming clusters in 2D space. These were assigned as the inputs for support vector machine (SVM) classification. H-scan imaging was able to differentiate between YUMM and YUMMER; YUMMER showed as more blue (higher frequency) in color. The H-scan parameters of blue fraction and SNR show significant differences between the two types of melanoma. Combining the four parameters, principal components formed clusters of YUMM and YUMMER. The clusters were separated by the decision planes obtained using SVM analysis with 88.5% classification accuracy. We anticipate clinical use of H-scan parameters for melanoma diagnosis.

22 November 2023 in Ultrasound in Medicine & Biology
Verified
Journal Article
Closed access
JBJihye Baek
Jihye Baek
SQShuyang S. Qin
Shuyang S. Qin
PPPeter A. Prieto
Peter A. Prieto
KPKevin Parker
Kevin Parker

23 December 2025 in IEEE Transactions on Ultrasonics
Verified
Journal Article
Closed access
HHHoda S. Hashemi
Hoda S. Hashemi
DHDongwoon Hyun
Dongwoon Hyun
NNNathan Nguyen
Nathan Nguyen
JBJihye Baek
Jihye Baek
ANArutselvan Natarajan
Arutselvan Natarajan
JDJeremy Dahl
Jeremy Dahl
Ultrasound molecular imaging (UMI) is an advanced imaging modality that shows promise in detecting cancer at early stages. It uses microbubbles as contrast agents, which are functionalized to bind to cancer biomarkers overexpressed on endothelial cells. A major challenge in UMI is isolating bound microbubble signal, which represents the molecular imaging signal, from that of free-floating microbubbles, which is considered background noise. In this work, we propose a fast GPU-based method using robust principal component analysis (RPCA) to distinguish bound microbubbles from free-floating ones. We explore the method using simulations and measure the accuracy using the Dice coefficient and RMS error as functions of the number of frames used in RPCA reconstruction. Experiments using stationary and flowing microbubbles in tissue-mimicking phantoms were used to validate the method. Additionally, the method was applied to data from ten transgenic mouse models of breast cancer development, injected with B7-H3 targeted microbubbles, and two mice injected with non-targeted microbubbles. The results showed that RPCA using 20 frames achieved a Dice score of 0.95 and a computation time of 0.2 seconds, indicating that 20 frames is potentially suitable for real-time implementation. On in vivo data, RPCA using 20 frames achieved a Dice score of 0.82 when compared with a conventional preclinical method, Differential Targeted Enhancement (DTE), indicating good agreement between the two, given the limitations of each method.

27 April 2026 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
JBJihye Baek
Jihye Baek
DHDongwoon Hyun
Dongwoon Hyun
ANArutselvan Natarajan
Arutselvan Natarajan
FTFarbod Tabesh
Farbod Tabesh
RPRamasamy Paulmurugan
Ramasamy Paulmurugan
JDJeremy Dahl
Jeremy Dahl
Ultrasound molecular imaging (USMI) is an imaging approach that utilizes targeted microbubbles (MBs) to highlight biomarkers of disease. While differential targeted enhancement (DTE) is the current state-of-the-art for USMI, its reliance on destructive pulses hinders real-time clinical application. We have developed a neural network-based nondestructive USMI, validated in vivo using a transgenic mouse model of spontaneous breast cancer. To enhance training efficacy despite a limited animal number (N=14), we utilized several augmentation strategies including the use of several targeted MB types for each animal to generate independent image and texture patterns, alternative DTE approaches (sham and injection DTE), and random patch selection, overall resulting in a total of 15,350 patches to train the network. The resulting nondestructive USMI produces an image of the pixelwise MB classification score of the presence of targeted MBs. Our nondestructive USMI achieved a correlation coefficient of 0.954 with DTE, a continuous dice coefficient of 0.863 for a molecular signal coverage of the lesion over 20%, and a higher AUC than DTE ( 0.954 vs. 0.845 ) compared to the reference image developed from the contrast enhanced ultrasound (CEUS) image and manual lesion contour. Nondestructive imaging during continuous motion of the transducer under elevation sweeps yielded fewer artifacts and higher AUC than DTE ( 0.953 vs. 0.892 ), compared to the reference image. This demonstrates the potential of free-hand and real-time nonde-structive imaging. Overall, nondestructive imaging showed comparable performance to DTE under stationary conditions and superior performance to DTE under transducer motion, indicating its clinical imaging potential.