A5003671567- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Advanced Neural Network Applications
- Prostate Cancer Diagnosis and Treatment
- Cell Image Analysis Techniques
- Artificial Intelligence in Healthcare
- Time Series Analysis and Forecasting
- Advanced Image and Video Retrieval Techniques
- Digital Imaging for Blood Diseases
- Generative Adversarial Networks and Image Synthesis
- Machine Learning and Data Classification
- Video Surveillance and Tracking Methods
- Surgical Simulation and Training
- Infrared Target Detection Methodologies
- Advanced X-ray and CT Imaging
- Advanced SAR Imaging Techniques
- Domain Adaptation and Few-Shot Learning
- Advanced Vision and Imaging
- Medical Imaging and Analysis
- MRI in cancer diagnosis
- Video Coding and Compression Technologies
- UAV Applications and Optimization
- Model Reduction and Neural Networks
- Image and Video Quality Assessment
- Anomaly Detection Techniques and Applications
University of Southern California
2021-2025
Southern California University for Professional Studies
2021
Information Technologies Institute
2019
Centre for Research and Technology Hellas
2019
University of Thessaly
2015
This paper presents the second edition of "drone-vs-bird" detection challenge, launched within activities 16-th IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS). The challenge's goal is to detect one or more drones appearing at some point in video sequences where birds may be also present, together with motion background foreground. Submitted algorithms should raise an alarm provide a position estimate only when drone while not issuing alarms birds, nor...
The popularity of Unmanned Aerial Vehicles (UAVs) is increasing year by and reportedly their applications hold great shares in global technology market. Yet, since UAVs can be also used for illegal actions, this raises various security issues that needs to encountered. Towards end, UAV detection systems have emerged detect further anticipate inimical drones. A very significant factor the maximum range which system's senses "see" an upcoming UAV. For those employ optical cameras detecting...
Clinically significant prostate cancer (csPCa) is a leading cause of death in men, yet it has high survival rate if diagnosed early. Bi-parametric MRI (bpMRI) reading become prominent screening test for csPCa. However, this process false positive (FP) rate, incurring higher diagnostic costs and patient discomfort. This paper introduces RadHop-Net, novel lightweight CNN FP reduction. The pipeline consists two stages: Stage 1 employs data driven radiomics to extract candidate ROIs. In...
While extensive research has been conducted on evalu-ating generative models, little the quality assessment and enhancement of individual-generated samples. We propose a lightweight generaliz-able evaluation framework, designed to evaluate en-hance models generated Our framework trains classifier-based dataset-specific model, enabling its application unseen extending compatibility with both deep learning ef-ficient machine learning-based methods. three novel metrics aiming at capturing...
You have accessJournal of UrologyCME1 Apr 2023MP09-05 AUTOMATED PROSTATE GLAND AND ZONES SEGMENTATION USING A NOVEL MRI-BASED MACHINE LEARNING FRAMEWORK CREATION OF SOFTWARE INTERFACE FOR USERS ANNOTATION Masatomo Kaneko, GIovanni E. Cacciamani, Yijing Yang, Vasileios Magoulianitis, Jintang Xue, Jiaxin Jinyuan Liu, Maria Sarah L. Lenon, Passant Mohamed, Darryl H. Hwang, Karan Gill, Manju Aron, Vinay Duddalwar, Suzanne Palmer, C.-C. Jay Kuo, Andre Luis Abreu, Inderbir and Chrysostomos Nikias...
Based on PixelHop and PixelHop++, which are recently developed using the successive subspace learning (SSL) framework, we propose an enhanced solution for object classification, called E-PixelHop, in this work. E-PixelHop consists of following steps. First, to decouple color channels a image, apply principle component analysis project RGB three onto two subspaces processed separately classification. Second, address importance multi-scale features, conduct pixel-level classification at each...
High Efficiency Video Coding (HEVC) is the new video compression standard, reducing bitrates nearly at half compared to H.264, offering potentially significant power savings for wireless transmission network interface. This reduction in bitrate achieved by a series of computationally expensive algorithms, thus making imperative optimize HEVC decoding order provide low-power implementation that can be used mobile devices. Extending Instruction Set Architecture (ISA) configurable...
You have accessJournal of UrologyCME1 Apr 2023MP55-20 A NOVEL MACHINE LEARNING FRAMEWORK FOR AUTOMATED DETECTION OF PROSTATE CANCER LESIONS CONFIRMED ON MRI-INFORMED TARGET BIOPSY Masatomo Kaneko, Giovanni E. Cacciamani, Vasileios Magoulianitis, Yijing Yang, Jintang Xue, Jiaxin Jinyuan Liu, Maria Sarah L. Lenon, Passant Mohamed, Darryl H. Hwang, Karan Gill, Manju Aron, Vinay Duddalwar, Suzanne Palmer, C.-C. Jay Kuo, Inderbir Andre Luis Abreu, and Chrysostomos Nikias KanekoMasatomo Kaneko...
Despite prolific work on evaluating generative models, little research has been done the quality evaluation of an individual generated sample. To address this problem, a lightweight sample (LGSQE) method is proposed in work. In training stage LGSQE, binary classifier trained real and synthetic samples, where data are labeled by 0 1, respectively. inference stage, assigns soft labels (ranging from to 1) each The value label indicates level; namely, better if its closer 0. LGSQE can serve as...
Automatic prostate segmentation is an important step in computer-aided diagnosis of cancer and treatment planning. Existing methods are based on deep learning models which have a large size lack transparency essential for physicians. In this paper, new data-driven 3D method MRI proposed, named PSHop. Different from methods, the core methodology PSHop feed-forward encoder-decoder system successive subspace (SSL). It consists two modules: 1) encoder: fine to coarse unsupervised representation...
Prostate Cancer is one of the most frequently occurring cancers in men, with a low survival rate if not early diagnosed. PI-RADS reading has high false positive rate, thus increasing diagnostic incurred costs and patient discomfort. Deep learning (DL) models achieve segmentation performance, although require large model size complexity. Also, DL lack feature interpretability are perceived as ``black-boxes" medical field. PCa-RadHop pipeline proposed this work, aiming to provide more...
You have accessJournal of UrologySurgical Technology & Simulation: Artificial Intelligence III (PD36)1 May 2024PD36-11 A NOVEL LIGHTWEIGHT MACHINE LEARNING MODEL FOR AUTOMATED RECLASSIFICATION OF THE INDEX LESION ON BIPARAMETRIC PROSTATE MRI Giovanni E. Cacciamani, Masatomo Kaneko, Vasileios Magoulianitis, Jiaxin Yang, Yijing Jintang Xue, Jinyuan Liu, Passant Mohamed, Darryl H. Hwang, Karanvir Gill, Lorenzo Storino Ramacciotti, Divyangi Paralkar, Manju Aron, Vinay Duddalwar, Suzanne L....
An unsupervised data-driven nuclei segmentation method for histology images, called CBM, is proposed in this work. CBM consists of three modules applied a block-wise manner: 1) color transform energy compaction and dimension reduction, 2) binarization, 3) incorporation geometric priors with morphological processing. comes from the first letter - "Color transform", "Binarization" "Morphological processing". Experiments on MoNuSeg dataset validate effectiveness method. outperforms all other...
A statistical attention localization (SAL) method is proposed to facilitate the object classification task in this work. SAL consists of three steps: 1) preliminary window selection via decision statistics, 2) map refinement, and 3) rectangular region finalization. computes soft-decision scores local squared windows uses them identify salient regions Step 1. To accommodate various sizes shapes, refines result obtain an more flexible shape 2. Finally, yields a using refined bounding box...
Among the existing modalities for 3D action recognition, flow has been poorly examined, although conveying rich motion information cues human actions. Presumably, its susceptibility to noise renders it intractable, thus challenging learning process within deep models. This work demonstrates use of sequence by a spatiotemporal model and further proposes an incremental two-level spatial attention mechanism, guided from skeleton domain, emphasizing features close body joint areas according...