Review
Surgery
Bishara Atiyeh, Saif Emsieh, Christopher Hakim, Rawad Chalhoub
Summary: Notoriously subjective, reporting aesthetic outcome in plastic surgery lacks scientific validation and relies on ill-defined endpoints and subjective measures. As the demand for aesthetic procedures increases, there is a need for better understanding of aesthetics and reliable, objective outcome measures. Advances in artificial intelligence (AI) offer potential for objective outcome analysis, such as facial emotions recognition systems, to quantify patients' reported outcomes and define success from the patients' perspective. This review aims to analyze the advantages and limitations of AI technology in objectively documenting aesthetic interventions' outcomes.
AESTHETIC PLASTIC SURGERY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Jung Hyun Yoon, Eun-Kyung Kim, Ga Ram Kim, Kyunghwa Han, Hee Jung Moon
Summary: This study assessed the impact of adding DBT or AI-CAD on recall rate and diagnostic performance in women undergoing mammographic surveillance after BCT. The results showed that adding DBT or AI-CAD reduced recall rates and improved diagnostic accuracy.
AMERICAN JOURNAL OF ROENTGENOLOGY
(2022)
Article
Engineering, Biomedical
Zoe Hu, Paola V. Nasute Fauerbach, Chris Yeung, Tamas Ungi, John Rudan, Cecil Jay Engel, Parvin Mousavi, Gabor Fichtinger, Doris Jabs
Summary: This study aims to develop a real-time automatic neural network-based tumor segmentation process for intraoperative guidance in breast-conserving surgery. The segmentation accuracy is evaluated using pixel-based metrics and expert visual rating. The results show that the neural networks provide consistent tumor segmentations that are well received by clinicians in the navigation system.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2022)
Article
Biochemistry & Molecular Biology
A. Stone, C. Kalahiki, L. Li, N. Hubig, F. Iuricich, H. Dunn
Summary: The aim of this research was to investigate the morphological differences of breast tumors between African American and Caucasian racial groups using machine learning and artificial intelligence methods. A supervised AI method was used to evaluate these differences, and significant variations were found in tumor and extracellular matrix (ECM) regions between the racial groups. Further analysis and characterization may provide new insight into the disparities associated with breast cancer incidence.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2023)
Article
Health Care Sciences & Services
Zhuang Wu, Hongkai Gu, Ronghua Hong, Ziwen Xing, Zhuoyu Zhang, Kangwen Peng, Yijing He, Ludi Xie, Jingxing Zhang, Yichen Gao, Yue Jin, Xiaoyun Su, Hongping Zhi, Qiang Guan, Lizhen Pan, Lingjing Jin
Summary: The study used a Kinect depth camera-based motion analysis system to quantify bradykinesia in Parkinson's disease (PD) and compared it with healthy control (HC) subjects. The results showed that PD patients had significantly lower frequencies and speeds of finger tapping, hand movement, hand pronation-supination movements, and leg agility compared to HCs. Combining kinematic features from different motor tasks improved the diagnostic value of differentiating PD patients from HCs.
Article
Health Care Sciences & Services
Roman Zeleznik, Jakob Weiss, Jana Taron, Christian Guthier, Danielle S. Bitterman, Cindy Hancox, Benjamin H. Kann, Daniel W. Kim, Rinaa S. Punglia, Jeremy Bredfeldt, Borek Foldyna, Parastou Eslami, Michael T. Lu, Udo Hoffmann, Raymond Mak, Hugo J. W. L. Aerts
Summary: The study evaluated the use of a deep-learning system for heart segmentation on CT scans in radiation oncology treatment planning. The system, trained with multi-center data and validated in a real-world dataset, showed improved segmentation time and agreement compared to manual methods. The results indicate that deep-learning algorithms can be successfully applied across medical specialties to enhance clinical care.
NPJ DIGITAL MEDICINE
(2021)
Article
Medicine, General & Internal
Timothy J. Whelan, Sally Smith, Sameer Parpia, Anthony W. Fyles, Anita Bane, Fei-Fei Liu, Eileen Rakovitch, Lynn Chang, Christiaan Stevens, Julie Bowen, Sawyna Provencher, Valerie Theberge, Anna Marie Mulligan, Zuzana Kos, Mohamed A. Akra, K. David Voduc, Tarek Hijal, Ian S. Dayes, Gregory Pond, James R. Wright, Torsten O. Nielsen, Mark N. Levine
Summary: Among women with T1N0 grade 1 or 2 luminal A breast cancer who had undergone breast-conserving surgery and received endocrine therapy, the incidence of local recurrence at 5 years was low without radiotherapy.
NEW ENGLAND JOURNAL OF MEDICINE
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Laura Kerschke, Stefanie Weigel, Alejandro Rodriguez-Ruiz, Nico Karssemeijer, Walter Heindel
Summary: The study evaluated the potential of artificial intelligence in discriminating between benign and malignant mammographic abnormalities, showing AI's ability to reduce false-positive rates and improve accuracy in recalling lesions. AI demonstrated the ability to decrease false-positives and non-FPR at the expense of a slight sensitivity reduction compared to human reading, especially in cases of mass-related lesions. Further prospective studies are needed to assess the practical benefits of AI in breast cancer screening.
EUROPEAN RADIOLOGY
(2022)
Article
Medicine, General & Internal
Ian H. H. Kunkler, Linda J. J. Williams, Wilma J. L. Jack, David A. A. Cameron, J. Michael Dixon
Summary: This study suggests that omission of radiotherapy in older women with low-risk, hormone receptor-positive early breast cancer may increase the risk of local recurrence, but has no detrimental effect on distant recurrence and overall survival.
NEW ENGLAND JOURNAL OF MEDICINE
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
I Skarping, M. Larsson, D. Fornvik
Summary: This proof of concept study investigated a deep learning-based method using digital mammograms to predict breast cancer patients' responses to neoadjuvant chemotherapy. The initial artificial intelligence model showed potential in aiding clinical decision-making. Further research, including method refinement and a larger sample size, is needed to explore the clinical utility of AI in predicting responses to neoadjuvant chemotherapy for breast cancer.
EUROPEAN RADIOLOGY
(2022)
Article
Psychology, Clinical
Luciano Boquete, Maria-Jose Vicente, Juan-Manuel Miguel-Jimenez, Eva-Maria Sanchez-Morla, Miguel Ortiz, Maria Satue, Elena Garcia-Martin
Summary: This study aims to identify objective biomarkers of fibromyalgia (FM) by using artificial intelligence algorithms to analyze structural data on the neuroretina. The results show significant differences in certain regions of the retinal nerve fiber layer (RNFL) and ganglion cell layers (GCL+) and (GCL++) in FM patients. The diagnostic aid system with an automatic classifier achieves high accuracy and discriminant capacity.
INTERNATIONAL JOURNAL OF CLINICAL AND HEALTH PSYCHOLOGY
(2022)
Article
Health Care Sciences & Services
Alessandra Panico, Gianluca Gatta, Antonio Salvia, Graziella Di Grezia, Noemi Fico, Vincenzo Cuccurullo
Summary: Breast cancer is the most common non-skin cancer in women and is influenced by habits and heredity. Regular screening, particularly through mammography, is crucial for early detection and increased chances of survival. Innovative techniques using artificial intelligence, such as radiomics, have shown promise in improving the quality of diagnosis for breast cancer.
JOURNAL OF PERSONALIZED MEDICINE
(2023)
Article
Oncology
Chengzhen Bao, Jie Shen, Yue Zhang, Yan Zhang, Wei Wei, Ziteng Wang, Jia Ding, Lili Han
Summary: This study aimed to evaluate the diagnostic performance of radiologists on breast cancer with artificial intelligence (AI) support. The results showed that radiologists with AI support demonstrated improved sensitivity, agreement rate, and Kappa value, with shorter reading time.
Review
Medicine, General & Internal
Akriti Nanda, Jesse Hu, Sarah Hodgkinson, Sanah Ali, Richard Rainsbury, Pankaj G. Roy
Summary: The evidence is uncertain about the oncological outcomes of OBCS compared to other surgical options for breast cancer. OBCS may have benefits in terms of reducing re-excision rates, but may also lead to more complications and higher recall rates. Further research is needed to evaluate the safety and efficacy of OBCS.
COCHRANE DATABASE OF SYSTEMATIC REVIEWS
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Anna Palmisano, Davide Vignale, Edda Boccia, Alessandro Nonis, Chiara Gnasso, Riccardo Leone, Marco Montagna, Valeria Nicoletti, Antonello Giuseppe Bianchi, Stefano Brusamolino, Andrea Dorizza, Marco Moraschini, Rahul Veettil, Alberto Cereda, Marco Toselli, Francesco Giannini, Marco Loffi, Gianluigi Patelli, Alberto Monello, Gianmarco Iannopollo, Davide Ippolito, Elisabetta Maria Mancini, Gianluca Pontone, Luigi Vignali, Elisa Scarnecchia, Mario Iannacone, Lucio Baffoni, Massimiliano Sperandio, Caterina Chiara de Carlini, Sandro Sironi, Claudio Rapezzi, Luca Antiga, Veronica Jagher, Clelia Di Serio, Cesare Furlanello, Carlo Tacchetti, Antonio Esposito
Summary: This study aims to develop an AI platform for risk stratification of COVID-19 patients based on clinical data and CT automatic analysis. Through rigorous model selection and data extraction, an automated risk scoring system was established. The system showed good predictive performance on two independent datasets.
Article
Oncology
C. Sessa, J. Balmana, S. L. Bober, M. J. Cardoso, N. Colombo, G. Curigliano, S. M. Domchek, D. G. Evans, D. Fischerova, N. Harbeck, C. Kuhl, B. Lemley, E. Levy-Lahad, M. Lambertini, J. A. Ledermann, S. Loibl, K. -A. Phillips, S. Paluch-Shimon
ANNALS OF ONCOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Eduardo Castro, Pedro M. Ferreira, Ana Rebelo, Isabel Rio-Torto, Leonardo Capozzi, Mafalda Falcao Ferreira, Tiago Goncalves, Tome Albuquerque, Wilson Silva, Carolina Afonso, Ricardo Gamelas Sousa, Claudio Cimarelli, Nadia Daoudi, Gabriel Moreira, Hsiu-yu Yang, Ingrid Hrga, Javed Ahmad, Monish Keswani, Sofia Beco
Summary: Every year, the VISUM summer school holds a competition for participants to learn and exchange knowledge about Computer Vision and Machine Learning in the context of fashion. This year's focus was on applying these methodologies to the fashion industry, which has shown rapid growth in e-commerce. The competition aimed to foster research and development in fashion outfit complementary product retrieval, and introduced a new fashion outfit dataset for benchmarking. The collaboration between academia and industry in this joint project has contributed to disseminating science and technology, promoting economic and social development, and connecting early-career researchers to real-world industry challenges.
MACHINE VISION AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Eduardo Castro, Jose Costa Pereira, Jaime S. Cardoso
Summary: Breast cancer is the most common and lethal cancer in women. Efforts have been made to develop accurate neural network-based computer-aided diagnosis systems for screening and prediction of this disease. This study presents a symmetry-based regularization method to improve the model's generalization ability to unseen examples in different environments.
MEDICAL IMAGE ANALYSIS
(2023)
Article
Pathology
Diana Montezuma, Sara P. Oliveira, Pedro C. Neto, Domingos Oliveira, Ana Monteiro, Jaime S. Cardoso, Isabel Macedo-Pinto
Summary: In this paper, the authors describe their experience in training machine learning models for AI applications in pathology, which often requires extensive annotation by human experts. They provide a simple and practical guide addressing annotation strategies for AI development in computational pathology, covering team interaction, ground-truth quality assessment, annotation types, and available software and hardware options. This guide aims to assist pathologists, researchers, and AI developers in the annotation process.
Correction
Multidisciplinary Sciences
Wilson Silva, Tiago Goncalves, Kirsi Haermae, Erich Schroeder, Verena Carola Obmann, Maria Cecilia Barroso, Alexander Poellinger, Mauricio Reyes, Jaime S. Cardoso
SCIENTIFIC REPORTS
(2023)
Article
Oncology
E. -A. Bonci, J. Correia Anacleto, M. -J. Cardoso
Summary: Simple breast conservation surgery (sBCS) has advanced to oncoplastic breast procedures (OBP) to improve psychosocial well-being and cosmetic outcome. Despite the increase in OBP use, the choice between sBCS and OBP is difficult. Research on their safety, efficacy, and patient-reported outcomes is lacking.
Article
Chemistry, Analytical
Ricardo Cruz, Diana Teixeira e Silva, Tiago Goncalves, Diogo Carneiro, Jaime S. Cardoso
Summary: Semantic segmentation involves classifying each pixel based on a set of classes. Conventional models have equal focus on classifying easy and hard-to-segment pixels, which is inefficient. This study proposes a framework where the model first produces a rough segmentation and then refines patches of the image estimated as difficult to segment. The framework is evaluated on four datasets and four state-of-the-art architectures, showing a fourfold acceleration in inference time with some tradeoff in output quality.
Article
Multidisciplinary Sciences
Sara P. Oliveira, Diana Montezuma, Ana Moreira, Domingos Oliveira, Pedro C. Neto, Ana Monteiro, Joao Monteiro, Liliana Ribeiro, Sofia Goncalves, Isabel M. Pinto, Jaime S. Cardoso
Summary: Cervical cancer is a common female cancer and leading cause of cancer-related death in women. Detecting pre-cancerous lesions is crucial, and machine learning models can assist pathologists in this task. A weakly-supervised methodology for grading cervical dysplasia is proposed, achieving a balanced accuracy of 71.07% and sensitivity of 72.18%.
SCIENTIFIC REPORTS
(2023)
Article
Biotechnology & Applied Microbiology
Nuno Freitas, Daniel Silva, Carlos Mavioso, Maria J. J. Cardoso, Jaime S. S. Cardoso
Summary: Breast cancer conservative treatment (BCCT) is commonly used for early breast cancer patients, removing the tumor and surrounding tissue while keeping healthy tissue intact. However, there is no gold-standard for evaluating the aesthetic results of BCCT. This paper proposes an improvement to the current technique for detecting breast contours in digital photographs, using a novel neural network solution.
BIOENGINEERING-BASEL
(2023)
Correction
Health Care Sciences & Services
Jelena Maksimenko, Pedro Pereira Rodrigues, Miki Nakazawa-Miklasevica, David Pinto, Edvins Miklasevics, Genadijs Trofimovics, Janis Gardovskis, Fatima Cardoso, Maria Joao Cardoso
JMIR FORMATIVE RESEARCH
(2023)
Article
Multidisciplinary Sciences
Orit Kaidar-Person, Marilia Antunes, Jaime D. Cardoso, Oriana Ciani, Helena Cruz, Rosa Di Micco, Oreste Gentilini, Tiago Goncalves, Pedro Gouveia, Jorg Heil, Pawel Kabata, Daniela Lopes, Marta Martinho, Henrique P. Martins, Carlos Mavioso, Martin Mika, Helena Montenegro, Helder Oliveira, Andre Pfob, Nicole Rotmensz, Timo Schinkothe, Giovani Silva, Rosana Tarricone, Maria-Joao Cardoso, CINDERELLA Consortium
Summary: The CINDERELLA trial aims to evaluate the impact of an artificial-intelligence cloud-based platform on patient education prior to therapy for breast cancer. The platform will provide patients with a visual representation of their potential aesthetic outcomes and an objective evaluation. The goal is to improve patient satisfaction, enhance psychosocial well-being, and reduce the need for additional surgeries.
Proceedings Paper
Computer Science, Artificial Intelligence
Helena Montenegro, Wilson Silva, Jaime S. Cardoso
Summary: The lack of interpretability in Deep Learning models is a barrier to their adoption in clinical settings. However, providing medical cases as explanations raises privacy concerns. To address this, a generative adversarial network is proposed to anonymize medical images while preserving explanatory evidence. The model is able to generate counterfactual explanations as well.
MEDICAL APPLICATIONS WITH DISENTANGLEMENTS, MAD 2022
(2023)
Article
Oncology
Pedro Filipe Pereira Gouveia, Rogelio Luna, Francisco Fontes, David Pinto, Carlos Mavioso, Joao Anacleto, Rafaela Timoteo, Joao Santinha, Tiago Marques, Fatima Cardoso, Maria Joao Cardoso
Summary: Augmented Reality (AR) has the potential to bring numerous benefits and educational applications in the virtual health ecosystem. This paper focuses on the future applications of AR in breast surgery education, specifically discussing two potential use cases and the technical requirements to make them possible.
Review
Surgery
Zoe Chia, Rachel X. N. Lee, Maria J. Cardoso, Kwok Leung Cheung, Ruth M. Parks
Summary: This review study found a lower uptake of oncoplastic breast surgery in older women compared to younger women. Further research is needed to understand the reasons behind this disparity.
BRITISH JOURNAL OF SURGERY
(2023)