Article
Pathology
Junhwi Kim, Naofumi Tomita, Arief A. Suriawinata, Saeed Hassanpour
Summary: This study evaluates the feasibility of deep learning models for classifying colorectal lesions into different categories. The model achieved high accuracy, sensitivity, and specificity on internal and external test sets, indicating its potential to assist pathologists in grading colorectal dysplasia and detecting adenocarcinoma. The model's high sensitivity on the external test set suggests its generalizability across different institutions.
AMERICAN JOURNAL OF PATHOLOGY
(2023)
Article
Gastroenterology & Hepatology
Quirine E. W. van der Zander, Ramon M. Schreuder, Roger Fonolla, Thom Scheeve, Fons van der Sommen, Bjorn Winkens, Patrick Aepli, Bu'Hussain Hayee, Andreas B. Pischel, Milan Stefanovic, Sharmila Subramaniam, Pradeep Bhandari, Peter H. N. de With, Ad A. M. Masclee, Erik J. Schoon
Summary: A computer-aided diagnosis system (CADx) utilizing artificial neural networks achieved significantly higher diagnostic accuracy for optical diagnosis of colorectal polyps compared with expert and novice endoscopists. The system had a diagnostic accuracy of 88.3% using HDWL images and 86.7% using BLI images, which was improved to 95.0% when combining both modalities. However, the use of the BLI Adenoma Serrated International Classification (BASIC) did not increase the diagnostic accuracy of endoscopists compared to intuitive optical diagnosis.
Article
Medicine, Research & Experimental
Ava Slotman, Minqi Xu, Katherine Lindale, Celine Hardy, Dan Winkowski, Regan Baird, Lina Chen, Priti Lal, Theodorus van der Kwast, Chelsea L. Jackson, Robert J. Gooding, David M. Berman
Summary: This study aimed to measure morphometric features relevant to grading criteria in noninvasive papillary urothelial carcinoma (NPUC) and build simplified classification models to distinguish between different grades objectively. The findings indicate that nuclear morphometry and automated mitotic figure counts can be used to objectively differentiate the grades of NPUC. These quantitative elements have the potential to revolutionize pathologic assessment and improve the prognostic utility of grade.
LABORATORY INVESTIGATION
(2023)
Article
Multidisciplinary Sciences
Dengbo Ji, Jinying Jia, Xinxin Cui, Zhaowei Li, Aiwen Wu
Summary: Mucinous colorectal adenocarcinoma (MC) has poor response to chemotherapy and prognosis compared to non-MC (NMC). Fibroblast activation protein (FAP) is upregulated in MC patients and negatively correlated with prognosis and therapeutic outcomes in CRC patients. FAP promotes CRC cell growth, invasion, metastasis, and chemoresistance. Myosin phosphatase Rho-interacting protein (MPRIP) is a direct interacting protein of FAP. FAP influences chemotherapy efficiency and prognosis by promoting crucial CRC functions and inducing recruitment of tumor-associated macrophages (TAMs) and M2 polarization through the Rho/Hippo/YAP signaling pathway. Knockdown of FAP reverses tumorigenicity and chemoresistance in CRC cells. FAP may serve as a marker for prognosis and therapeutic target to overcome chemoresistance in MC patients.
Article
Oncology
Yeonu Choi, Jaehong Aum, Se-Hoon Lee, Hong-Kwan Kim, Jhingook Kim, Seunghwan Shin, Ji Yun Jeong, Chan-Young Ock, Ho Yun Lee
Summary: The study developed a deep learning model for predicting high-grade patterns in lung adenocarcinomas and assessing their prognostic performance in advanced lung cancer patients who underwent neoadjuvant or definitive CCRT. The model showed an area under the curve value of 0.8 and successfully stratified survival in an independent validation set. Patients with a high probability of Micropapillary or Solid patterns (MPSol) estimated by the DL model had a significantly higher risk of death. The DL model can be useful in estimating high-grade histologic patterns in lung adenocarcinomas and predicting clinical outcomes of advanced lung cancer patients undergoing neoadjuvant or definitive CCRT.
Article
Oncology
Yu-Wen Zhou, Yi-Xiu Long, Ye Chen, Ji-Yan Liu, Dan Pu, Jia-Yan Huang, Feng Bi, Qiu Li, Hong-Feng Gou, Meng Qiu
Summary: The study compared the efficacy of first-line bevacizumab plus chemotherapy and cetuximab plus chemotherapy in metastatic colorectal cancer patients with mucinous adenocarcinoma or mucinous component. The results showed that bevacizumab-based chemotherapy was associated with significantly better overall survival in patients with mucinous adenocarcinoma or mucinous component, regardless of tumor sites.
Article
Oncology
Hao Fu, Weiming Mi, Boju Pan, Yucheng Guo, Junjie Li, Rongyan Xu, Jie Zheng, Chunli Zou, Tao Zhang, Zhiyong Liang, Junzhong Zou, Hao Zou
Summary: PDAC, one of the deadliest cancer types, requires histopathology image analysis for detection and diagnosis. Manual diagnosis is time-consuming and lacks accuracy, leading to the development of deep-learning methods as an alternative to feature extraction-based classification. The proposed model in this study demonstrates high accuracy in diagnosing PDAC using histopathological images.
FRONTIERS IN ONCOLOGY
(2021)
Article
Medicine, General & Internal
Eunjue Yi, Naoki Sunaguchi, Jeong Hyeon Lee, Chul-Yong Kim, Sungho Lee, Sanghoon Jheon, Masami Ando, Yangki Seok
Summary: The study evaluated the clinical implication of synchrotron radiation imaging techniques for human lung adenocarcinoma in comparison with pathologic examination. The research found that synchrotron radiation could provide high-resolution images of lung adenocarcinoma which were highly correlated with images from pathologic examinations.
Article
Ecology
Abdelfattah El Moussaoui, Mohammed Bourhia, Fatima Zahra Jawhari, Hind Khalis, Mohamed Chedadi, Abdelkrim Agour, Ahmad Mohammad Salamatullah, Abdulhakeem Alzahrani, Heba Khalil Alyahya, Asdaf Alotaibi, Dalila Bousta, Amina Bari
Summary: The study aimed to investigate the response of Withania frutescens to changes in environmental conditions, analyzing aspects such as topography, climate, morphology, histology, and phytochemistry. The results showed that the plant exhibited changes in various aspects under different conditions, indicating that it has developed alternative strategies to cope with environmental changes.
FRONTIERS IN ECOLOGY AND EVOLUTION
(2021)
Article
Immunology
Mayank Baranwal, Santhoshi Krishnan, Morgan Oneka, Timothy Frankel, Arvind Rao
Summary: Early detection of pancreatic ductal adenocarcinoma (PDAC) is crucial to prevent metastatic spread. Current automated grading methods rely on accurate identification of cell features or spatially informed indices, but do not provide insights into accurate disease grade identification.
FRONTIERS IN IMMUNOLOGY
(2021)
Article
Biology
Junyong Shen, Yan Hu, Xiaoqing Zhang, Yan Gong, Ryo Kawasaki, Jiang Liu
Summary: This paper proposes a novel Structure-Oriented Transformer (SoT) framework to construct the relationship between lesions and retina on clinical datasets, using the self-attention mechanism. Extensive experiments demonstrate the effectiveness of SoT components and its superiority in neovascular Age-related Macular Degeneration (nAMD) grading. The algorithm also shows good generality on a publicly available retinal diseases dataset. Overall, it receives a rating of 8 out of 10.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Medicine, General & Internal
Masayuki Tsuneki, Fahdi Kanavati
Summary: Poorly differentiated colorectal adenocarcinoma is associated with poor prognosis, and histopathological diagnosis from endoscopic biopsy specimens can benefit from AI tools to aid pathologists. Deep learning models were trained to classify poorly differentiated colorectal ADC from Whole Slide Images, achieving high accuracy in test cases.
Article
Computer Science, Artificial Intelligence
Trinh Thi Le Vuong, Kyungeun Kim, Boram Song, Jin Tae Kwak
Summary: This study introduces a joint categorical and ordinal learning framework to improve cancer grading in pathology image analysis. By utilizing a new loss function, the contrast between correctly classified and misclassified examples is enhanced.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Computer Science, Artificial Intelligence
Yi Lin, Jingchi Jiang, Dongxin Chen, Zhaoyang Ma, Yi Guan, Xiguang Liu, Haiyan You, Jing Yang
Summary: This study proposes an acne diagnosis method called Diagnostic Evidence Distillation (DED) that can adapt to the characteristics of acne diagnosis and be applied under different acne criteria. The DED framework utilizes convolutional neural networks (CNNs) and a teacher-student structure to improve diagnosis performance. Experimental results show that DED outperforms current state-of-the-art methods and reaches the diagnostic level of dermatologists.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Oncology
Shenlun Chen, Meng Zhang, Jiazhou Wang, Midie Xu, Weigang Hu, Leonard Wee, Andre Dekker, Weiqi Sheng, Zhen Zhang
Summary: This study developed an automated approach based on histology whole-slide images to evaluate the glandular formation density in colorectal cancer patients, which can serve as a prognostic factor. The results showed that the deep survival grade had improved discrimination in survival prediction.
FRONTIERS IN ONCOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
I. Khurram, M. M. Fraz, M. Shahzad, N. M. Rajpoot
Summary: The Dense-CaptionNet is a region-based deep architecture proposed for fine-grained description of image semantics, which localizes and describes each object/region in the image separately to generate a more detailed scene description. The network consists of three components working together to generate region descriptions, object relationships, and object attributes, feeding into a sentence generation module for generating a grammatically correct, single-line description of the whole scene using an encoder-decoder formulation.
COGNITIVE COMPUTATION
(2021)
Review
Pathology
Lisa Browning, Richard Colling, Emad Rakha, Nasir Rajpoot, Jens Rittscher, Jacqueline A. James, Manuel Salto-Tellez, David R. J. Snead, Clare Verrill
Summary: The measures to control COVID-19 outbreak will likely continue until a suitable vaccine or treatment is found, impacting clinical services and research, with pathologists working remotely and changes to protocols and studies. Digital pathology and artificial intelligence may play a key role in safeguarding clinical services and pathology-based research in the current and future climate.
JOURNAL OF CLINICAL PATHOLOGY
(2021)
Article
Oncology
Ian A. Cree, Blanca Iciar Indave Ruiz, Jiri Zavadil, James McKay, Magali Olivier, Zisis Kozlakidis, Alexander J. Lazar, Chris Hyde, Stefan Holdenrieder, Ros Hastings, Nasir Rajpoot, Arnaud de la Fouchardiere, Brian Rous, Jean Claude Zenklusen, Nicola Normanno, Richard L. Schilsky
Summary: The gaps in translating research findings into clinical practice, particularly in cancer diagnosis and management, have been recognized for decades. The WHO Classification of Tumours provides valuable international standards for cancer diagnosis and plays a crucial role in evidence synthesis and standard setting. To address the challenges in translating research findings to tumour classification, the International Collaboration for Cancer Classification and Research (ICR-R-3) has been established to coordinate efforts in standard setting and best practice recommendations.
INTERNATIONAL JOURNAL OF CANCER
(2021)
Review
Pathology
Ayesha S. Azam, Islam M. Miligy, Peter K-U Kimani, Heeba Maqbool, Katherine Hewitt, Nasir M. Rajpoot, David R. J. Snead
Summary: Digital pathology (DP) has the potential to revolutionize histopathology practice by improving efficiency and diagnostic accuracy. However, limited evidence of reliability has been a barrier to wider adoption. This meta-analysis showed equivalent performance of DP compared to light microscopy for routine diagnosis, with specific areas of diagnostic discrepancy identified.
JOURNAL OF CLINICAL PATHOLOGY
(2021)
Article
Ophthalmology
Amun Sachdev, Anshu Sachdev, Susan P. Mollan, David Snead, Harpreet S. Ahluwalia
Summary: This study reports a case of total anterior staphyloma resulting from untreated acanthamoeba keratitis, highlighting the importance of patient compliance in the management of fungal and acanthamoeba keratitis and in preventing vision-threatening sequelae.
EYE & CONTACT LENS-SCIENCE AND CLINICAL PRACTICE
(2022)
Article
Pathology
Nehal M. Atallah, Michael S. Toss, Clare Verrill, Manuel Salto-Tellez, David Snead, Emad A. Rakha
Summary: This study examined the impact of missing tissue on breast whole slide images (WSI) and found a negative linear correlation between the frequency of missing tissue and scanning time/image file size. Quality control measures improved image quality and reduced WSI failure rates, with missing tissue having little diagnostic consequence.
Article
Pathology
Noorul Wahab, Islam M. Miligy, Katherine Dodd, Harvir Sahota, Michael Toss, Wenqi Lu, Mostafa Jahanifar, Mohsin Bilal, Simon Graham, Young Park, Giorgos Hadjigeorghiou, Abhir Bhalerao, Ayat G. Lashen, Asmaa Y. Ibrahim, Ayaka Katayama, Henry O. Ebili, Matthew Parkin, Tom Sorell, Shan E. Ahmed Raza, Emily Hero, Hesham Eldaly, Yee Wah Tsang, Kishore Gopalakrishnan, David Snead, Emad Rakha, Nasir Rajpoot, Fayyaz Minhas
Summary: This paper addresses the importance of annotations in Computational Pathology (CPath) projects and the current lack of well-defined guidelines. By presenting a large-scale annotation exercise, the authors provide annotation guidelines and best practice recommendations for CPath projects.
JOURNAL OF PATHOLOGY CLINICAL RESEARCH
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ellhia Sudin, Mitchell Searjeant, George Partridge, Peter Phillips, Louise Hiller, David Snead, Ian Ellis, Yan Chen
Summary: The study found a significant association between pathologists' experience and reading times, number of fixations, and diagnostic accuracy. Greater experience is linked to stronger cognitive and visual processing capabilities, as well as a greater use of magnification changes.
JOURNAL OF MEDICAL IMAGING
(2022)
Article
Hematology
Sarah Gooding, Naser Ansari-Pour, Mohammad Kazeroun, Kubra Karagoz, Ann Polonskaia, Mirian Salazar, Evie Fitzsimons, Korsuk Sirinukunwattana, Selina Chavda, Maria Ortiz Estevez, Fadi Towfic, Erin Flynt, William Pierceall, Daniel Royston, Kwee Yong, Karthik Ramasamy, Paresh Vyas, Anjan Thakurta
Summary: The acquisition of a multidrug refractory state is a major cause of mortality in myeloma. Previous studies have found a relationship between cereblon protein (CRBN) and immunomodulatory drugs (IMiDs) resistance, but no genetic mutations associated with IMiDs resistance have been identified. This study identified two genes, COPS7B and COPS8, in the chromosome region 2q37, which are associated with IMiDs resistance. This region may serve as a potential clinical marker for IMiDs resistance.
Review
Oncology
Mohsin Bilal, Mohammed Nimir, David Snead, Graham S. Taylor, Nasir Rajpoot
Summary: Immunotherapy, particularly in the context of colorectal cancer, has seen significant advancements and the integration of artificial intelligence (AI) to predict biomarkers has been explored. Microsatellite instability (MSI) is a popular biomarker in this field, and AI has been utilized to predict MSI status and tumor mutation burden. This survey also discusses the potential of AI in predicting immune cell infiltrates and utilizing alternate data modalities such as immunohistochemistry and gene expression. The future directions emphasize the promise of AI in accelerating exploration and benefiting patients in the field of immunotherapy for colorectal cancer.
BRITISH JOURNAL OF CANCER
(2023)
Article
Pathology
Mahmoud Ali, Harriet Evans, Peter Whitney, Fayyaz Minhas, David R. J. Snead
Summary: The archiving of whole slide images is a challenge in digital pathology implementation due to the large amount of data generated. By examining the combination of SNOMED codes, it is possible to identify which cases are likely to be recalled, thus reducing the number of archived cases.
JOURNAL OF CLINICAL PATHOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Simon Graham, Quoc Dang Vu, Mostafa Jahanifar, Shan E. Ahmed Raza, Fayyaz Minhas, David Snead, Nasir Rajpoot
Summary: The recent advances in deep learning have greatly improved the performance of image analysis for digitised pathology slides. However, these deep models are often trained for a single task and require a large amount of training data. In this paper, a multi-task learning approach is proposed to address these issues by leveraging data from multiple sources and achieving simultaneous prediction for different tasks. The learned representation is also shown to be transferable for additional tasks.
MEDICAL IMAGE ANALYSIS
(2023)
Article
Oncology
Shorouk Makhlouf, Noorul Wahab, Michael Toss, Asmaa Ibrahim, Ayat G. Lashen, Nehal M. Atallah, Suzan Ghannam, Mostafa Jahanifar, Wenqi Lu, Simon Graham, Nigel P. Mongan, Mohsin Bilal, Abhir Bhalerao, David Snead, Fayyaz Minhas, Shan E. Ahmed Raza, Nasir Rajpoot, Emad Rakha
Summary: In this study, artificial intelligence (AI) was used to assess the prognostic significance of tumour infiltrating lymphocytes (TILs) in a large cohort of luminal breast cancer patients. The results showed that TILs counts and their spatial distribution had predictive value for the prognosis of early-stage breast cancer patients.
BRITISH JOURNAL OF CANCER
(2023)