Review
Biochemistry & Molecular Biology
Avan Kader, Julia Brangsch, Jan O. Kaufmann, Jing Zhao, Dilyana B. Mangarova, Jana Moeckel, Lisa C. Adams, Ingolf Sack, Matthias Taupitz, Bernd Hamm, Marcus R. Makowski
Summary: This review summarizes recent developments in molecular imaging markers for magnetic resonance imaging of prostate cancer, aiming to improve the molecular characterization of the tumor and enable targeted therapies to suppress tumor growth or reduce tumor size in a non-invasive manner.
Article
Food Science & Technology
Tanmay Sarkar, Tanupriya Choudhury, Nikunj Bansal, V. R. Arunachalaeshwaran, Mars Khayrullin, Mohammad Ali Shariati, Jose Manuel Lorenzo
Summary: Food adulteration is a serious concern that affects community health. Red chilli powder is a commonly used ingredient worldwide, but it is often found to be adulterated with brick powder. This study is the first attempt to use machine learning algorithms to detect adulteration in red chilli powder. The researchers created a dataset of high-quality images showing red chilli powder adulterated with different proportions of red brick powder. They applied various image color space filters and used mean and histogram feature extraction techniques. The best classification model was the Cat Boost classifier with HSV color space histogram features, and the best regression model was the Extra Tree regressor with Lab color space histogram features.
FOOD ANALYTICAL METHODS
(2023)
Article
Computer Science, Information Systems
Yuejing Qian, Zengyou Zhang, Bo Wang
Summary: Prostate cancer is a challenging malignant tumor to detect accurately. The newly designed method ProCDet based on MR images shows competitive performance in efficiently and accurately detecting prostate cancer.
Article
Radiology, Nuclear Medicine & Medical Imaging
Mark A. Anderson, Sarah Mercaldo, Ryan Chung, Ethan Ulrich, Randall W. Jones, Mukesh Harisinghani
Summary: This study aimed to evaluate the improvement in detection of clinically significant prostate cancer by adding a computer-aided diagnostic (CAD) generated MRI series. The results showed that the CAD-generated MRI series significantly improved the diagnostic performance and inter-reader agreement.
ACADEMIC RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Yue Lin, Mason J. Belue, Enis C. Yilmaz, Stephanie A. Harmon, Julie An, Yan Mee Law, Lindsey Hazen, Charisse Garcia, Katie M. Merriman, Tim E. Phelps, Nathan S. Lay, Antoun Toubaji, Maria J. Merino, Bradford J. Wood, Sandeep Gurram, Peter L. Choyke, Peter A. Pinto, Baris Turkbey
Summary: This study examined the impact of image quality on prostate cancer detection using an artificial intelligence algorithm. The results showed that higher quality T2-weighted images had a better clinically significant cancer detection rate in targeted biopsy for PI-RADS 4 lesions.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Patricia M. Johnson, Angela Tong, Awani Donthireddy, Kira Melamud, Robert Petrocelli, Paul Smereka, Kun Qian, Mahesh B. Keerthivasan, Hersh Chandarana, Florian Knoll
Summary: The study utilized a variational network (VN) for image reconstruction to accelerate prostate MRI exams, showing that diagnostic biparametric prostate MRI exams can be performed in less than 4 minutes using deep learning methods, potentially enabling rapid screening prostate MRI.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2022)
Article
Medicine, General & Internal
I-Ling Chen, Yen-Jen Wang, Chang-Cheng Chang, Yu-Hung Wu, Chih-Wei Lu, Jia-Wei Shen, Ling Huang, Bor-Shyh Lin, Hsiu-Mei Chiang
Summary: Individuals with dark skin types are more prone to pigmentary disorders, with melasma being particularly difficult to treat and often recurring. Objective measurement of melanin amount is important in evaluating treatment response, with the use of full-field optical coherence tomography and a computer-aided detection system proving to be effective in extracting various melanin features.
Article
Physiology
Zhenglin Yi, Zhenyu Ou, Jiao Hu, Dongxu Qiu, Chao Quan, Belaydi Othmane, Yongjie Wang, Longxiang Wu
Summary: This study evaluated a new deep neural network-based computer-aided diagnosis method for automatic localization and classification of prostate cancer. The results showed that the method achieved high performance in prostate cancer localization and classification.
FRONTIERS IN PHYSIOLOGY
(2022)
Article
Chemistry, Multidisciplinary
Sarmad Maqsood, Robertas Damasevicius, Rytis Maskeliunas
Summary: This paper presents a deep learning system for breast cancer screening, using mammogram images to achieve early detection of breast cancer. The proposed approach includes modified contrast enhancement, transferable texture convolutional neural network (TTCNN), and energy layer, as well as deep features and convolutional sparse image decomposition to improve classification accuracy. Experimental results demonstrate that the proposed method outperforms previous methods in classifying mammogram images.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Electrical & Electronic
Dong Liu, Long Wang, Yu Du, Ming Cong, Yongyao Li
Summary: In this study, a real-time and accurate automatic detection and segmentation algorithm for 3-D MR and TRUS images of the prostate is proposed, which achieves efficient and accurate detection and segmentation even under poor image quality conditions. The experimental results on public and private datasets demonstrate the superiority of this method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Review
Biology
Kosmia Loizidou, Rafaella Elia, Costas Pitris
Summary: Cancer, particularly breast cancer, is a significant global health concern and major cause of morbidity and mortality. Mammography is effective for early detection and management, but accurately identifying and interpreting breast lesions is challenging. Computer-Aided Diagnosis (CAD) systems have been developed to assist radiologists in detecting and classifying breast cancer. This review examines recent literature on the use of both traditional feature-based machine learning and deep learning algorithms for automatic detection and classification of breast cancer in mammograms, as well as FDA-approved CAD systems and potential future opportunities in this field.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Andrea Ponsiglione, Arnaldo Stanzione, Gianluigi Califano, Marco De Giorgi, Claudia Colla Ruvolo, Imma D'Iglio, Simone Morra, Nicola Longo, Massimo Imbriaco, Renato Cuocolo
Summary: The purpose of this study was to assess the impact of prostate MRI image quality on the identification of extraprostatic extension of disease. The results showed that image quality had little impact on the accuracy of EPE Grade and Likert Scale Score, but had some impact on the MSKCCn score. The study suggests that assessing image quality using PI-QUAL is important for accurately evaluating EPE.
EUROPEAN JOURNAL OF RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Giorgio Brembilla, Salvatore Lavalle, Tom Parry, Michele Cosenza, Tommaso Russo, Elio Mazzone, Francesco Pellegrino, Armando Stabile, Giorgio Gandaglia, Alberto Briganti, Francesco Montorsi, Antonio Esposito, Francesco De Cobelli
Summary: The impact of Prostate Imaging Quality (PI-QUAL) scores on the diagnostic performance of multiparametric MRI (mpMRI) in a targeted biopsy cohort was investigated. The results showed that scans of suboptimal quality (PI-QUAL < 4) were associated with lower positive predictive value (PPV) for clinically significant prostate cancer (csPCa) and lower detection rates in both PI-RADS 3 and PI-RADS 4-5.
EUROPEAN JOURNAL OF RADIOLOGY
(2023)
Article
Medicine, General & Internal
Valentina Giannini, Simone Mazzetti, Giovanni Cappello, Valeria Maria Doronzio, Lorenzo Vassallo, Filippo Russo, Alessandro Giacobbe, Giovanni Muto, Daniele Regge
Summary: This study evaluated the diagnostic performance of CAD system for non-experienced readers in detecting and characterizing Prostate Cancer (PCa). The results showed that CAD assistance significantly increased sensitivity without negatively affecting specificity, while also reducing overall reporting time significantly.
Article
Computer Science, Artificial Intelligence
Debendra Muduli, Rakesh Ranjan Kumar, Jitesh Pradhan, Abhinav Kumar
Summary: Early detection and diagnosis are crucial for decreasing breast cancer mortality rate in medical image analysis. Extreme learning machine (ELM), a randomized learning technique, plays a vital role in learning feed-forward networks with fast learning speed and good generalization. The performance of the ELM model is evaluated on the standard ultra-sound breast cancer dataset, BUSI, and compared with other variants of ELM.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Biochemical Research Methods
Leonardo Ayala, Fabian Isensee, Sebastian J. Wirkert, Anant S. Vemuri, Klaus H. Maier-Hein, Baowei Fei, Lena Maier-Hein
Summary: Multispectral imaging technique provides valuable information on tissue composition. Real-time application of this technique in interventional medicine is challenging due to long acquisition times for hyperspectral data. This study proposes a new approach to band selection using Monte Carlo simulations, which can reproduce or even improve the results of spectral measurement with a small subset of bands. The investigation also explores domain adaptation techniques to address the domain shift caused by simulations. Preliminary results suggest that 10-20 bands are sufficient to closely reproduce spectral measurements with 101 bands in the 500-700 nm range. The domain adaptation technique, which only requires unlabeled in vivo measurements, outperforms the pure in silico band selection method.
BIOMEDICAL OPTICS EXPRESS
(2022)
Proceedings Paper
Engineering, Biomedical
Ka'Toria Leitch, Martin Halicek, Maysam Shahedi, James Little, Amy Y. Chen, Baowei Fei
Summary: This study utilizes hyperspectral imaging (HSI) and radiomics to extract radiomic features of papillary thyroid carcinoma (PTC) specimens, and successfully predicts tumor aggressiveness through shape features. The HSI-based radiomic method provides a useful tool for oncologists to determine tumors with intermediate to high risk and make clinical decisions.
MEDICAL IMAGING 2022: COMPUTER-AIDED DIAGNOSIS
(2022)
Proceedings Paper
Engineering, Biomedical
Ka'Toria Leitch, Maysam Shahedi, James D. Dormer, Quyen N. Do, Yin Xi, Matthew A. Lewis, Christina L. Herrera, Catherine Y. Spong, Ananth J. Madhuranthakam, Diane M. Twickler, Baowei Fei
Summary: This study uses MRI images to predict the presence of PAS and the need for cesarean hysterectomy. By analyzing the placenta and uterus regions of interest and incorporating information from their dilation, the study achieves high accuracy in predicting hysterectomy and classifying suspected PAS.
MEDICAL IMAGING 2022: COMPUTER-AIDED DIAGNOSIS
(2022)
Proceedings Paper
Engineering, Biomedical
James Huang, Junyu Guo, Ivan Pedrosa, Baowei Fei
Summary: Respiratory motion is a major source of error in quantitative analysis of MRI data. This study proposes a deep learning approach to correct motion effects and applies it to kidney MRI imaging applications.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
Ted Shi, Maysam Shahedi, Kayla Caughlin, James D. Dormer, Ling Ma, Baowei Fei
Summary: This study introduces a semi-automated, deep learning-based approach to cardiac segmentation, which achieved good segmentation results by selecting points to mimic user interaction and training a fully convolutional neural network. The method showed promising performance for chamber-by-chamber delineation of the heart in CT images.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
Jeff Young, Maysam Shahedi, James D. Dormer, Brett A. Johnson, Jeffrey Gahan, Baowei Fei
Summary: In this study, PVC-plasticizer and silicone rubbers were used to create realistic and durable kidney phantoms with contrast under both ultrasound and X-ray imaging. The radiodensity properties of PVC-based gels were characterized to allow adjustable image intensity and contrast. By using a two-part molding process, internal structures of the kidney were created for greater customization. The study found that PVC exhibited good contrast under X-ray imaging and performed excellently for ultrasound imaging, making it a more suitable material for kidney phantoms compared to silicone. The durability and shelf life of the PVC-based phantoms were also observed to be superior to commonly used agar-based phantoms.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
Patric Bettati, James D. Dormer, Maysam Shahedi, Baowei Fei
Summary: Ultrasound-guided biopsy is commonly used for disease detection and diagnosis. This study aims to improve the localization of suspicious lesions by registering preoperative imaging with real-time ultrasound imaging and using augmented reality technology. The preliminary results show the feasibility of using multiple imaging modalities in an augmented reality-guided system for ultrasound-guided prostate biopsy.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Meeting Abstract
Obstetrics & Gynecology
Christina L. Herrera, Meredith J. Kim, Quyen N. Do, David M. Owen, Baowei Fei, Ananth J. Madhuranthakam, Yin Xi, Matthew A. Lewis, Diane M. Twickler, Catherine Y. Spong
AMERICAN JOURNAL OF OBSTETRICS AND GYNECOLOGY
(2022)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Chuangye Wan, Ling Ma, Xiabi Liu, Baowei Fei
Summary: The study proposes a computer-aided method for accurate classification of lung nodules using expert knowledge, achieving high accuracy and AUC values on benchmark dataset. The method has great potential for various applications in lung cancer detection, diagnosis, and therapy.
MEDICAL IMAGING 2021: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2021)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Ling Ma, Maysam Shahedi, Ted Shi, Martin Halicek, James Little, Amy Y. Chen, Larry L. Myers, Baran D. Sumer, Baowei Fei
Summary: In this study, a fully convolutional network (FCN) was utilized to classify tumors and assess margins on hyperspectral images of SCC, achieving high accuracy in pixel-level tissue classification. The evaluated tumor margins in most patients were within a small distance, and the classification time was quick.
MEDICAL IMAGING 2021: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2021)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Lei Tao, Ling Ma, Maoqiang Xie, Xiabi Liu, Zhiqiang Tian, Baowei Fei
Summary: This study proposes an automatic segmentation method for the prostate on MR images based on anatomy knowledge, achieving a high segmentation accuracy in experiments.
MEDICAL IMAGING 2021: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2021)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Patric Bettati, James D. Dormer, Jeff Young, Maysam Shahedi, Baowei Fei
Summary: Cardiac catheterization is a delicate strategy with risks, so we developed a virtual reality simulation and visualization method to reduce risks, shorten operation time, and improve accuracy.
MEDICAL IMAGING 2021: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2021)
Proceedings Paper
Engineering, Biomedical
Ka'Toria Edwards, Avneesh Chhabra, James Dormer, Phillip Jones, Robert D. Boutin, Leon Lenchik, Baowei Fei
Summary: CT is commonly used for monitoring muscle changes in patients, with manual segmentation of CT slices being a time-consuming task. A CNN-based segmentation method is proposed in this study, allowing for automatic segmentation of abdominal muscles and reducing the time required for obtaining relevant information.
MEDICAL IMAGING 2020: BIOMEDICAL APPLICATIONS IN MOLECULAR, STRUCTURAL, AND FUNCTIONAL IMAGING
(2021)
Proceedings Paper
Engineering, Biomedical
Martin Halicek, Samuel Ortega, Himar Fabelo, Carlos Lopez, Marylene Lejaune, Gustavo M. Callico, Baowei Fei
Summary: A method utilizing a generative adversarial network to synthesize hyperspectral imaging from standard RGB images of breast cancer tissue has been proposed in this study. The synthesized hyperspectral images achieved high structural similarity and low mean absolute error compared to actual hyperspectral images. Further research is needed to validate the effectiveness of this method on larger datasets.
MEDICAL IMAGING 2020: DIGITAL PATHOLOGY
(2021)
Proceedings Paper
Engineering, Biomedical
Ling Ma, Martin Halicek, Baowei Fei
Summary: This study investigates the feasibility of using wavelet-based features from hyperspectral images for head and neck cancer detection, showing that the proposed feature yields better classification accuracy and that using different types and orders of mother wavelets leads to different classification results.
MEDICAL IMAGING 2020: BIOMEDICAL APPLICATIONS IN MOLECULAR, STRUCTURAL, AND FUNCTIONAL IMAGING
(2021)
