Article
Engineering, Biomedical
Aswathy Rajendra Kurup, Jeff Wigdahl, Jeremy Benson, Manel Martinez-Ramon, Peter Soliz, Vinayak Joshi
Summary: Cerebral malaria (CM) is a fatal syndrome commonly observed in children under 5 in Sub-saharan Africa and Asia. The presence of retinal lesions known as malarial retinopathy (MR), including whitening and hemorrhages, is highly specific to CM. Over-diagnosis of CM occurs in up to 23% of cases due to similar clinical symptoms with pneumonia or meningitis, leading to untreated conditions and potential death or disability. A low-cost and high-specificity diagnostic technique based on transfer learning (TL) has been developed to detect CM, achieving a 96% specificity using inexpensive retinal cameras.
BIOCYBERNETICS AND BIOMEDICAL ENGINEERING
(2023)
Review
Computer Science, Interdisciplinary Applications
Barkha Kakkar, Mohit Goyal, Prashant Johri, Yogesh Kumar
Summary: Artificial Intelligence is used in detecting malaria to provide accurate diagnosis with minimal human interference. This paper aims to create an AI-based system for detecting and classifying malaria parasites in microscopic images. Different deep learning models are trained and evaluated for their performance in terms of precision, loss, F1 score, recall, and accuracy. MobileNetV2 achieves the highest accuracy, while ResNet152V2 has the best loss value of 0.005. DenseNet121 achieves perfect precision, recall, and F1 score of 1.00. The models are also evaluated using a confusion matrix to compute actual and predicted values for different classes.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Multidisciplinary Sciences
Carla Parra, Felipe Grijalva, Bryan Nunez, Alejandra Nunez, Noel Perez, Diego Benitez
Summary: Captive environments lead to the spread and multiplication of parasites among different reptile species, weakening their immune response. We propose an approach that uses convolutional neural networks and transfer learning to detect and classify parasites, and it has been validated on a stool image dataset, achieving high accuracy.
Article
Engineering, Electrical & Electronic
Jiajia Liao, Yingchao Piao, Jinhe Su, Guorong Cai, Xingwang Huang, Long Chen, Zhaohong Huang, Yundong Wu
Summary: This article proposed an unsupervised cluster guided detection framework to address the issue of limited detection ability in scenes where objects are densely distributed in high-resolution aerial images. Experimental results show that the method outperforms existing baseline methods in two popular aerial image datasets.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Review
Biochemistry & Molecular Biology
Bugra Oezdemir, Ralf Reski
Summary: Cytoskeletal filaments play crucial roles in biological cells and organisms, and understanding their geometric and topological organization is key for revealing their functions. High-resolution microscopy and sophisticated image processing software are required for accurate segmentation. Recent advancements in deep learning have started to simplify this task.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2021)
Article
Multidisciplinary Sciences
Taehan Koo, Moon Hwan Kim, Mihn-Sook Jue
Summary: This study focuses on detecting fungal infections more quickly, conveniently, and consistently through deep learning using images obtained from real-world practice. The research aims to develop an automatic hyphae detection system with high sensitivity and specificity for practical clinical applications.
Article
Geochemistry & Geophysics
Liang Zhang, Jianda Cheng, Jiafei Liu, Tao Liu, Deliang Xiang, Yi Su
Summary: This letter presents an unsupervised ship detection method in SAR images using superpixel segmentation and cross stage partial network (CSPNet). The method achieves pixel-level detection map instead of bounding box result.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Automation & Control Systems
Xingzhen Wu, Guojun Mao, Shuli Xing
Summary: This study proposes a novel normalizing flow model called AFlow to improve image-based anomaly detection. By using multi-scale convolutions and an attention module, the proposed method achieves more accurate prediction results and higher parameter efficiency compared to previous models.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Menghui Niu, Kechen Song, Liming Huang, Qi Wang, Yunhui Yan, Qinggang Meng
Summary: A novel unsupervised stereoscopic saliency detection method based on a binocular line-scanning system is proposed, which can simultaneously acquire high-precision image and profile information while avoiding decoding distortion. Experimental results show that the method outperforms 15 state-of-the-art algorithms.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Geochemistry & Geophysics
Caijun Ren, Xiangyu Wang, Jian Gao, Xiren Zhou, Huanhuan Chen
Summary: This study introduces a novel change detection framework utilizing Generative Adversarial Network (GAN) to generate better coregistered images, improving the performance of change detection algorithms. Experimental results demonstrate that this method is less sensitive to the issue of unregistered images and effectively utilizes deep learning structures.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Mamadou Dian Bah, Adel Hafiane, Raphael Canals
Summary: Crop row detection is crucial for smart farming, and deep learning approaches have been widely studied for this task. However, the requirement of large labeled datasets poses a challenge, especially in agriculture where labeling data is tedious and expensive. Graph-based unsupervised techniques offer a promising alternative by incorporating structured information such as plant relationships.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Geochemistry & Geophysics
Yuan Zhou, Xiangrui Li, Keran Chen, Sun-Yuan Kung
Summary: This article introduces an unsupervised progressive learning framework (UPLF) for optical aerial image change detection. The proposed method addresses the problems of ignoring spatial information and introducing new errors in existing unsupervised change detection techniques. By using original change maps as labeled samples and applying progressive learning, the proposed method achieves more reliable labeling and accurate detection results. Experimental results demonstrate the highly competitive performance of the proposed method.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Chemistry, Analytical
Elzbieta Kubera, Agnieszka Kubik-Komar, Pawel Kurasinski, Krystyna Piotrowska-Weryszko, Magdalena Skrzypiec
Summary: This research evaluated the automatic analysis of pollen material based on digital microscopic photos using a deep neural network called YOLO. The results showed that YOLO outperformed other deep learning methods in terms of detection and recognition, with mean average precision ranging from 86.8% to 92.4% for the test sets. The study also identified difficulties in correctly classifying pollen material, such as the similarities and overlapping of grains.
Article
Chemistry, Analytical
Kyungkoo Jun
Summary: This study focuses on the rectification of vehicle plate images to improve the accuracy of license-plate recognition. A perspective-transformation process is used to project the images as if taken from the front geometrically. A heatmap-based corner-detection model is proposed for unsupervised domain adaptation, which allows the model to work well on license plates from different countries.
Article
Radiology, Nuclear Medicine & Medical Imaging
Byungjai Kim, Kinam Kwon, Changheun Oh, Hyunwook Park
Summary: An anomaly detection method for pixel-level detection in multicontrast MRI was proposed using a deep neural network. The algorithm showed significant improvements in quantitative and qualitative evaluations compared to previous methods. The effectiveness of each module in the proposed framework was validated through ablation studies.
Article
Hematology
Olga Mykhailova, Tracey R. Turner, Carly Olafson, Anita Howell, Susan N. Nahirniak, Juanita Wizniak, Hanan Y. N. Gerges, Troy Baldwin, Gwen Clarke, Jason P. Acker
Summary: Irradiation of RBCs can effectively inactivate T lymphocytes to prevent TA-GVHD. Leukoreduction and storage of blood products may be an alternative prevention strategy. Viability of residual WBCs in stored RBCs remains high, although their proliferative ability decreases over time.
Letter
Medicine, General & Internal
Willy Albert Flegel, Melanie Bodnar, Gwen Clarke, Judith Hannon, Lani Lieberman
CANADIAN MEDICAL ASSOCIATION JOURNAL
(2021)
Article
Hematology
Rebecca Cardigan, Helen New, Lise Estcourt, Eugene Zhiburt, Rounak Dubey, Jesper Bengtsson, Magnus Joud, Carlos Castillo, Joan Cid, Miquel Lozano, Dhana Gounder, Peter Flanagan, Sarah Morley, Gwen Clarke, Dana Devine, Salwa Hindawi, Aqeel AlOtaibi, Carolina Bonnet Bub, Jose Mauro Kutner, Toshiyuki Ikeda, Naoko Goto, Hitoshi Okazaki, Magali J. Fontaine, Jeremiah Pasion, Linda Song, Tom Latham, Jean-Louis Kerkhoffs, Masja de Haas, Jaap Jan Zwaginga, Birgit S. Gathof, Katharina Ommer, France Pirenne, Michel Raba, Anne Francois, James Daly, Tanya Powley, Nancy Dunbar
Article
Hematology
Rebecca Cardigan, Helen New, Lise Estcourt, Eugene Zhiburt, Rounak Dubey, Jesper Bengtsson, Magnus Joud, Carlos Castillo, Joan Cid, Miquel Lozano, Dhana Gounder, Peter Flanagan, Sarah Morley, Gwen Clarke, Dana Devine, Salwa Hindawi, Aqeel AlOtaibi, Carolina Bonnet Bub, Jose Mauro Kutner, Toshiyuki Ikeda, Naoko Goto, Hitoshi Okazaki, Magali J. Fontaine, Jeremiah Pasion, Linda Song, Tom Latham, Jean-Louis Kerkhoffs, Masja de Haas, Jaap Jan Zwaginga, Birgit S. Gathof, Katharina Ommer, France Pirenne, Michel Raba, Anne Francois, James Daly, Tanya Powley, Nancy Dunbar
Editorial Material
Hematology
Sunitha Vege, Aline Floch, Christine Lomas-Frances, Gwen Clarke, Connie M. Westhoff
Article
Hematology
Olga Gajic-Veljanoski, Chunmei Li, Alexis K. Schaink, Jennifer Guo, Nadine Shehata, George S. Charames, Barbra Vrijer, Gwen Clarke, Petros Pechlivanoglou, Nanette Okun, Rita Kandel, Joseph Dooley, Caroline Higgins, Vivian Ng, Nancy Sikich
Summary: This study examined the cost-effectiveness of noninvasive fetal RhD blood group genotyping in nonalloimmunized and alloimmunized pregnancies in Canada. The results showed that the genotyping was slightly associated with a higher probability of maternal alloimmunization in nonalloimmunized pregnancies, but resulted in a reduced number of Rh immunoglobulin injections. In alloimmunized pregnancies, the genotyping was found to be less expensive and more effective compared to usual care.
Review
Hematology
Omar Hajjaj, Gwen Clarke, Lani Lieberman
Summary: This study investigated the current status and practice of cord blood testing, revealing diversity and lack of standardization. It suggests that cord blood tests should only be performed when clinically indicated and provides 15 guidance statements.
Article
Hematology
Carly Olafson, Nishaka William, Anita Howell, Lynnette Beaudin, Balkar Gill, Gwen Clarke, Stephanie Stephens, Dora Lopes-Carvalho, Debra Lane, Peter Schubert, Ken McTaggart, Jason P. Acker
Summary: The present study evaluates the impact of preparation and storage of small-dose red cell concentrates (RCCs) on in vitro red cell quality. The results indicate that small-dose RCCs, whether irradiated or not, meet the quality criteria required for safe transfusion.
Article
Hematology
Marissa Laureano, Gwen Clarke, Matthew T. S. Yan
Summary: This paper outlines the Rare Blood Program of Canadian Blood Services (CBS) and provides data on its management of rare red cell requests and inventory. The provision of rare red cells involves multiple considerations, including communication, confirmation, recruitment, and logistics. The paper also suggests that new technologies may affect how rare donors are identified and recruited in the future.
Article
Hematology
Shuoyan Ning, Pierre-Aurele Morin, Allahna Elahie, Na Li, Yang Liu, Rebecca Barty, Gwen Clarke, Michelle Zeller, Nancy M. Heddle
Summary: This study assessed the impact of a transfusion policy for anti-K alloantibodies on females of child-bearing potential and evaluated the feasibility of a KEL1 negative transfusion policy. The results showed that transfusion was responsible for 25% of alloimmunizations in females with anti-K, and the current red blood cell inventory was sufficient to meet demand.
Article
Hematology
Robert Liwski, Gwen Clarke, Calvino Cheng, Syed Sibte Raza Abidi, Samina Raza Abidi, Jason George Quinn
Summary: This study successfully validates the application of flow cytometry in RBC phenotyping. Flow cytometry offers advantages such as high automation, quantitative assessment, and low reagent volumes compared to conventional methods. It can be used for high-throughput, low-cost phenotyping in blood suppliers and hospital BTS.
Article
Hematology
Shuang Niu, Megan Vetsch, Lynnette Beaudin, Melanie Bodnar, Gwen Clarke
Summary: This study compared the traditional manual SIAT antibody titration method with the newer automated ASP method and found that ASP titers were, on average, 1.33 dilutions higher than SIAT titers. The study suggests that a titer cutoff of >= 32 is appropriate for most clinically significant antibodies using ASP.
Meeting Abstract
Hematology
Terrie Butler-Foster, Aditi Khandelwal, Akash Gupta, Matthew T. S. Yan, Gwen Clarke
Meeting Abstract
Hematology
Carly Olafson, Lynnette Beaudin, Debra Jean Lane, Balkar Gill, Gwen Clarke, Stephanie Stephens, Dora Lopes-Carvalho, Jason P. Acker
Meeting Abstract
Hematology
Omar Ali Hajjaj, Gwen Clarke, Lani Lieberman
