Featured
-
-
Article
| Open AccessVOLTA: an enVironment-aware cOntrastive ceLl represenTation leArning for histopathology
While machine learning platforms can improve the assessment of Hematoxylin & Eosin (H&E) stained-tumour tissue images, current models typically require manual cell-type annotations in training. Here, the authors develop VOLTA, a self-supervised machine learning framework to improve cell representation learning in H&E images based on the cells environment
- Ramin Nakhli
- , Katherine Rich
- & Ali Bashashati
-
Article
| Open AccessA multicenter proof-of-concept study on deep learning-based intraoperative discrimination of primary central nervous system lymphoma
Correct diagnosis of primary central nervous system lymphoma is key in determining treatment, however, this depends on pathology analysis. Here, the authors develop a deep learning method to diagnose primary nervous system lymphoma from stained whole-slide images.
- Xinke Zhang
- , Zihan Zhao
- & Muyan Cai
-
Article
| Open AccessTeacher-student collaborated multiple instance learning for pan-cancer PDL1 expression prediction from histopathology slides
PDL1 expression is a common biomarker for immunotherapy response in cancer, and it is usually quantified using immunohistochemistry. Here, the authors develop a weakly supervised learning approach combining multiple instance learning and a teacher-student framework to predict PDL1 expression from histopathological imaging.
- Darui Jin
- , Shangying Liang
- & Xiangzhi Bai
-
Article
| Open AccessDeepDOF-SE: affordable deep-learning microscopy platform for slide-free histology
Histopathology can be limited by the time-consuming and labour-intensive preparation of slides from resected tissue. Here, the authors report DeepDOF-SE, a deep-learning-enabled microscope to rapidly scan intact tissue at cellular resolution without the need for physical sectioning.
- Lingbo Jin
- , Yubo Tang
- & Ashok Veeraraghavan
-
Article
| Open AccessIdentification of spatially-resolved markers of malignant transformation in Intraductal Papillary Mucinous Neoplasms
The current stratification of Intraductal Papillary Mucinous Neoplasms (IPMN) is based on clinical and histological features rather than molecular ones. Here, the authors use spatial transcriptomics to characterise IPMN patient samples, and identify markers associated with progression to pancreatic cancer.
- Antonio Agostini
- , Geny Piro
- & Carmine Carbone
-
Article
| Open AccessDevelopment and validation of an interpretable model integrating multimodal information for improving ovarian cancer diagnosis
Ovarian cancer diagnosis can be complex and can be improved through integration of multimodal data. Here, the authors develop OvcaFinder which can significantly outperform clinical models using ultrasound images and known clinical characteristics.
- Huiling Xiang
- , Yongjie Xiao
- & Hao Chen
-
Article
| Open AccessAcidity-activatable upconversion afterglow luminescence cocktail nanoparticles for ultrasensitive in vivo imaging
Activatable afterglow luminescence nanoprobes reduce unspecific signals and improve imaging fidelity, but their utility is limited by a requisition of donor-acceptor distance (>10 nm) in common biomarker-activatable designs. Here, the authors address this issue by developing organic afterglow luminescence cocktail nanoparticles for acid-activatable upconversion afterglow luminescence imaging.
- Yue Jiang
- , Min Zhao
- & Qingqing Miao
-
Article
| Open AccessEnhancing the fairness of AI prediction models by Quasi-Pareto improvement among heterogeneous thyroid nodule population
Artificial Intelligence (AI) models for medical diagnosis often face challenges of generalizability and fairness. Here, the authors show that the Quasi-Pareto Improvement approach is widely applicable to improving AI models among less-prevalent subgroups, promoting equitable healthcare outcomes.
- Siqiong Yao
- , Fang Dai
- & Hui Lu
-
Article
| Open AccessRegression-based Deep-Learning predicts molecular biomarkers from pathology slides
Cancer biomarkers are often continuous measurements, which poses challenges for their prediction using classification-based deep learning. Here, the authors develop a regression-based deep learning method to predict continuous biomarkers - such as the homologous repair deficiency score - from cancer histopathology images.
- Omar S. M. El Nahhas
- , Chiara M. L. Loeffler
- & Jakob Nikolas Kather
-
Article
| Open AccessMotility and tumor infiltration are key aspects of invariant natural killer T cell anti-tumor function
Invariant natural killer T (iNKT) cells are important contributors to anti-tumour immunity, but they often become dysfunctional in cancers. Here authors show that inhibited iNKT intra-tumour motility and iNKT cell exclusion from tumours by macrophages both contribute to their diminished function in cancer, and by therapeutic interference with the respective motility and iNKT-macrophage interaction pathways, their function can be restored.
- Chenxi Tian
- , Yu Wang
- & Li Bai
-
Article
| Open AccessA light-activatable theranostic combination for ratiometric hypoxia imaging and oxygen-deprived drug activity enhancement
Monitoring the level of hypoxia in a tumour is of use when treating by chemotherapy or photodynamic therapy, but can be challenging. Here, the authors report the development of a theranostic combination for light activated hypoxia imaging and modulation, and prodrug activation.
- Lei Ge
- , Yikai Tang
- & Xiqun Jiang
-
Article
| Open AccessHigh-throughput deconvolution of 3D organoid dynamics at cellular resolution for cancer pharmacology with Cellos
Computational methods to analyse 3D organoids in high-throughput and with high cellular resolution remain scarce. Here, the authors propose Cellos, a high-throughput pipeline for 3D organoid segmentation using classical algorithms and a trained convolutional neural network.
- Patience Mukashyaka
- , Pooja Kumar
- & Jeffrey H. Chuang
-
Article
| Open AccessDeep learning of cell spatial organizations identifies clinically relevant insights in tissue images
Cell spatial organization in tissue provides essential insights into diseases. Here, the authors show Ceograph, a graph convolutional network, for the analysis of pathology images to predict patient outcomes, highlighting cellular markers to guide personalized treatments and enhance biological understanding.
- Shidan Wang
- , Ruichen Rong
- & Guanghua Xiao
-
Article
| Open AccessSecretory GFP reconstitution labeling of neighboring cells interrogates cell–cell interactions in metastatic niches
Methodologies to study the mechanisms of cell–cell interactions in metastatic niches remain scarce. Here, the authors develop a secretory GFP reconstitution-based system to tag tissue-resident cells neighboring on cancer cells within the metastatic niche.
- Misa Minegishi
- , Takahiro Kuchimaru
- & Satoshi Nishimura
-
Article
| Open AccessSingle-cell morphological and topological atlas reveals the ecosystem diversity of human breast cancer
Whole-slide images (WSI) and digital pathology are valuable approaches for the analysis of tumours and their microenvironments. Here, the authors present scMTOP, a framework to characterise tumour ecosystems and intercellular relationships at the single-cell level from WSIs, which they apply to breast cancer samples.
- Shen Zhao
- , De-Pin Chen
- & Zhi-Ming Shao
-
Article
| Open AccessIntegrated radiogenomics models predict response to neoadjuvant chemotherapy in high grade serous ovarian cancer
Response to treatment in high grade serous ovarian carcinoma (HGSOC) is highly variable. Here, the authors leverage a radiogenomic model to predict neoadjuvant chemotherapy response in HGSOC, including clinical data, medical imaging, and blood-based biomarkers such as CA-125 and ctDNA features.
- Mireia Crispin-Ortuzar
- , Ramona Woitek
- & James D. Brenton
-
Article
| Open AccessGlycerol-weighted chemical exchange saturation transfer nanoprobes allow 19F/1H dual-modality magnetic resonance imaging-guided cancer radiotherapy
Radiotherapy (RT) sensitizers have been used to overcome tumor hypoxia and improve response to RT. Here the authors design and characterize a pH and oxygen sensitive nano-molecular probe for imaging-guided cancer radiotherapy.
- Rong A
- , Haoyu Wang
- & Xilin Sun
-
Article
| Open AccessNeuropathologist-level integrated classification of adult-type diffuse gliomas using deep learning from whole-slide pathological images
Determining glioma types directly from whole-slide images (WSIs) would be of great diagnostic utility. Here, the authors develop a deep learning model to identify diffuse glioma types from WSIs according to the 2021 WHO classification across multiple cohorts and with interpretable features.
- Weiwei Wang
- , Yuanshen Zhao
- & Zhenyu Zhang
-
Article
| Open AccessIntegrated molecular and multiparametric MRI mapping of high-grade glioma identifies regional biologic signatures
Glioma tumours are known to be heterogenous in mutation and gene expression patterns, but sampling limitations can lead to inaccurate detection of evolutionary events. Here, the authors carry out multi-omics analysis of multi-regional biopsies from 68 patients and show differential mutations in non-enhancing regions.
- Leland S. Hu
- , Fulvio D’Angelo
- & Nhan L. Tran
-
Article
| Open AccessDeep learning-enabled realistic virtual histology with ultraviolet photoacoustic remote sensing microscopy
Oncologic tumour resection is not fully accurate. Here the authors report a label-free virtual histological imaging method based on a non-contact, reflection-mode ultraviolet photoacoustic remote sensing and scattering microscope, combined with unsupervised deep learning using a cycle-consistent GAN.
- Matthew T. Martell
- , Nathaniel J. M. Haven
- & Roger J. Zemp
-
Article
| Open AccesspH-gated nanoparticles selectively regulate lysosomal function of tumour-associated macrophages for cancer immunotherapy
A high abundance of tumour associated macrophages with immunosuppressive properties is associated with inefficient anti-tumour immune responses. Here the authors report the design and characterization of pH-gated nanoparticles selectively targeting and reprogramming M2-like macrophages in the tumour microenvironment, re-sensitizing tumours to immune checkpoint blockade.
- Mingmei Tang
- , Binlong Chen
- & Yiguang Wang
-
Article
| Open AccessNon-invasive assessment of normal and impaired iron homeostasis in the brain
Assessment of different iron compounds in the living brain remains an open challenge. Here, the authors present a magnetic resonance imaging method which is sensitive to the iron homeostasis in the brain, and increases the detection of tumor tissue.
- Shir Filo
- , Rona Shaharabani
- & Aviv A. Mezer
-
Article
| Open AccessBiology-guided deep learning predicts prognosis and cancer immunotherapy response
The clinical application of data-driven deep learning models remains challenging. Here, a biology-guided deep learning approach allows the simultaneous prediction of the tumour immune and stromal microenvironment status as well as treatment outcomes from medical images in gastric cancer.
- Yuming Jiang
- , Zhicheng Zhang
- & Ruijiang Li
-
Article
| Open AccessEGFR-targeted fluorescence molecular imaging for intraoperative margin assessment in oral cancer patients: a phase II trial
By using tumor-specific fluorescent tracers, fluorescence molecular imaging (FMI) can be used to visualize tumor tissues with high specificity. Here the authors report the results of a phase II trial to evaluate the diagnostic accuracy of an EGFR-targeted FMI for intraoperative margin assessment in patients with oral squamous cell carcinoma.
- Jaron G. de Wit
- , Jasper Vonk
- & Max J. H. Witjes
-
Article
| Open AccessCombining amino acid PET and MRI imaging increases accuracy to define malignant areas in adult glioma
Magnetic Resonance Imaging (MRI) is normally used to define glioma boundaries for biopsies, surgery and radiotherapy. Here, the authors show that adding FET/PET imaging improves accuracy to define malignant areas of contrast-enhancing gliomas.
- Maciej Harat
- , Józefina Rakowska
- & Bogdan Małkowski
-
Article
| Open AccessVirtual alignment of pathology image series for multi-gigapixel whole slide images
The spatial organization of a tumor affects how it grows and responds to treatment. Here, the authors present VALIS, a software to align sets of whole slide images (WSI) with state-of-the-art accuracy, enabling spatial studies of the tumor ecology.
- Chandler D. Gatenbee
- , Ann-Marie Baker
- & Alexander R. A. Anderson
-
Article
| Open AccessCancer-associated fibroblast classification in single-cell and spatial proteomics data
Cancer-associated fibroblasts (CAFs) have different subtypes and play diverse roles in the tumour microenvironment. Here, the authors use single-cell RNA-seq and multiplex imaging mass cytometry data to propose a CAF classification scheme of nine subtypes across different cancer types.
- Lena Cords
- , Sandra Tietscher
- & Bernd Bodenmiller
-
Comment
| Open AccessTowards evidence-based response criteria for cancer immunotherapy
Early detection of immunotherapy-induced tumor response is of major benefit for patients but can be complicated by therapy-induced pseudoprogression. A consensus guideline-iRECIST- was developed as a modification of Response Evaluation Criteria in Solid Tumours (RECIST version 1.1). Here we describe which next steps are required to test its validity and how novel approaches for response criteria might be developed and included.
- Elena Garralda
- , Scott A. Laurie
- & Elisabeth G. E. de Vries
-
Article
| Open AccessA local water molecular-heating strategy for near-infrared long-lifetime imaging-guided photothermal therapy of glioblastoma
Neodymium (Nd)-doped nanoparticles have been described for imaging-guided photothermal therapy. Here the authors design a Nd-Yb co-doped nanomaterial as nearinfrared long-lifetime imaging-guided waterheating probe, showing photothermal ablation in a glioblastoma pre-clinical mode
- Dongkyu Kang
- , Hyung Shik Kim
- & Joonseok Lee
-
Article
| Open AccessBridging clinic and wildlife care with AI-powered pan-species computational pathology
Artificial Intelligence (AI) has the potential of assisting the study and diagnosis of veterinary cancers. Here, the authors build a cancer digital pathology atlas encompassing multiple animal species and demonstrate an AI approach for comparative pathology, which yields insights about immune response and morphological similarities.
- Khalid AbdulJabbar
- , Simon P. Castillo
- & Yinyin Yuan
-
Article
| Open AccessHistopathology images predict multi-omics aberrations and prognoses in colorectal cancer patients
Histopathological analysis is an essential tool in diagnosing colorectal cancer, but is limited in predicting prognosis and molecular profiles. Here, the authors designed a machine learning-based platform to predict multi-omics profiles and prognosis from pathology images.
- Pei-Chen Tsai
- , Tsung-Hua Lee
- & Kun-Hsing Yu
-
Article
| Open AccessNoncanonical amino acids as doubly bio-orthogonal handles for one-pot preparation of protein multiconjugates
Site-specific protein multi-conjugates are important for both scientific and translational research. Here, the authors genetically encode unnatural amino acids which contain both tetrazine and azide, and use the doubly bio-orthogonal handles to generate bi- and tri-conjugate proteins in high yields.
- Yong Wang
- , Jingming Zhang
- & Tao Liu
-
Article
| Open AccessA pH-responsive T1-T2 dual-modal MRI contrast agent for cancer imaging
Magnetic resonance imaging (MRI) is a well-established non-invasive medical imaging technology. Here, to improve the performance of the technique, the authors describe the design of a pH-responsive T1-T2 dual-modal MRI contrast agent, showing enhanced imaging sensitivity in preclinical cancer models.
- Hongwei Lu
- , An Chen
- & Leilei Tian
-
Article
| Open AccessA doxycycline- and light-inducible Cre recombinase mouse model for optogenetic genome editing
Achieving spatial control of gene expression is important. Here the authors report an optimised photoactivatable Cre recombinase system, doxycycline- and light-inducible Cre recombinase (DiLiCre), and generate a DiLiCre mouse line which they use for mutagenesis in vivo and positional cell-tracing.
- Miguel Vizoso
- , Colin E. J. Pritchard
- & Jacco van Rheenen
-
Article
| Open AccessSynthetic biology-instructed transdermal microneedle patch for traceable photodynamic therapy
An effective delivery system and imaging method for 5-Aminolevulinic acid (5-ALA)-based photodynamic therapy facilitated by the conversion of 5-ALA to protoporphyrin IX (PpIX) are lacking. Here, reversing the hypoxic tumour microenvironment can increase the in vivo biosynthesis of PpIX through the regulation of PpIX-related synthetases for traceable photodynamic therapy.
- Gang He
- , Yashi Li
- & Peng Huang
-
Article
| Open AccessMultiscale profiling of protease activity in cancer
The activity of multiple enzymes is dysregulated in cancer, but this cannot always be measured through enzyme expression. Here, the authors develop methods to measure protease activity across the organism, tissue, and single cell scales, and identify protease dysregulation in lung cancer and its response to targeted therapy.
- Ava P. Amini
- , Jesse D. Kirkpatrick
- & Sangeeta N. Bhatia
-
Article
| Open AccessNear-infrared phosphorescent carbon dots for sonodynamic precision tumor therapy
Combining sonodynamic properties and NIR fluorescence into a single material is desired for deep tissue applications. Here, the authors report on carbon dot sono-sensitizers engineered with a narrow bandgap and coated with cancer cell membrane for targeted NIR guided sonodynamic cancer therapy.
- Bijiang Geng
- , Jinyan Hu
- & Longxiang Shen
-
Article
| Open AccessAdversarial attacks and adversarial robustness in computational pathology
Artificial Intelligence can support diagnostic workflows in oncology, but they are vulnerable to adversarial attacks. Here, the authors show that convolutional neural networks are highly susceptible to white- and black-box adversarial attacks in clinically relevant classification tasks.
- Narmin Ghaffari Laleh
- , Daniel Truhn
- & Jakob Nikolas Kather
-
Article
| Open AccessPhosphorylcholine-conjugated gold-molecular clusters improve signal for Lymph Node NIR-II fluorescence imaging in preclinical cancer models
Fluorescent tracers facilitate the identification and subsequent collection of tumour draining lymph node biopsies, enabling important clinical assessment. Here, the authors present a molecular gold nanocluster NIR-II fluorescent imaging probe and demonstrate its utility to visualise draining lymph nodes in breast and colon cancer mouse models.
- Ani Baghdasaryan
- , Feifei Wang
- & Hongjie Dai
-
Article
| Open AccessIn vivo tumor immune microenvironment phenotypes correlate with inflammation and vasculature to predict immunotherapy response
Standard assessment of immune infiltration of biopsies is not sufficient to accurately predict response to immunotherapy. Here, the authors show that reflectance confocal microscopy can be used to quantify dynamic vasculature and inflammatory features to better predict treatment response in skin cancers.
- Aditi Sahu
- , Kivanc Kose
- & Milind Rajadhyaksha
-
Article
| Open AccessTraject3d allows label-free identification of distinct co-occurring phenotypes within 3D culture by live imaging
There are currently a lack of tools to detect heterogeneity in 3D cultures. Here the authors report Traject3d as a framework to identify heterogeneous states in 3D culture and to understand how these give rise to distinct phenotypes using label-free multi-day time-lapse imaging.
- Eva C. Freckmann
- , Emma Sandilands
- & David M. Bryant
-
Article
| Open AccessNear-infrared II plasmonic porous cubic nanoshells for in vivo noninvasive SERS visualization of sub-millimeter microtumors
In vivo surface-enhanced Raman scattering (SERS) imaging allows non-invasive visualization of tumours for biomedical applications. Here, the authors report porous cubic AuAg alloy nanoshells exhibiting plasmonic properties and porosity-dependant SERS in the second window of the near-infrared for in vivo tumour detection.
- Linhu Li
- , Renting Jiang
- & Ming Li
-
Article
| Open AccessNoninvasive imaging of the tumor immune microenvironment correlates with response to immunotherapy in gastric cancer
Tumour microenvironment has been linked with immunotherapy response in gastric cancer. Here, the authors use CT-based radiomics to predict neutrophils-to-lymphocyte ratio and response to immunotherapy.
- Weicai Huang
- , Yuming Jiang
- & Guoxin Li
-
Article
| Open AccessThe spatio-temporal evolution of multiple myeloma from baseline to relapse-refractory states
Spatially and temporally resolved data can improve our understanding of evolution and treatment resistance in multiple myeloma (MM). Here, the authors analyse spatial and longitudinal heterogeneity in MM patients using multi-region sequencing, and identify subclones associated with relapse and therapy resistance.
- Leo Rasche
- , Carolina Schinke
- & Niels Weinhold
-
Article
| Open AccessInstant diagnosis of gastroscopic biopsy via deep-learned single-shot femtosecond stimulated Raman histology
Diagnosis of gastric cancer currently requires gastroscopic biopsy, which requires time and expertize to perform. Here, the authors demonstrate a femto-SRS imaging method which showed high accuracy in diagnosing gastric cancer without the need for pathologistbased diagnosis.
- Zhijie Liu
- , Wei Su
- & Minbiao Ji
-
Article
| Open AccessSpatiotemporal analysis of glioma heterogeneity reveals COL1A1 as an actionable target to disrupt tumor progression
It is essential to improve our understanding of the features that influence aggressiveness and invasion in high grade gliomas (HGG). Here, the authors characterize dynamic anatomical structures in HGG called oncostreams, which are associated with tumor growth and are regulated by COL1A1.
- Andrea Comba
- , Syed M. Faisal
- & Pedro R. Lowenstein
-
Article
| Open AccessBioorthogonally activatable cyanine dye with torsion-induced disaggregation for in vivo tumor imaging
Expanding the responsive dyes repertoire is currently a developing field in biorthogonal chemistry. In this article, the authors develop fluorophores that turn on their near-infrared fluorescence upon biorthogonal reaction based on a “torsion-induced disaggregation” approach, allowing for sensitive in vivo imaging of tumors.
- Xianghan Zhang
- , Jingkai Gao
- & Zhongliang Wang
-
Article
| Open AccessFast raster-scan optoacoustic mesoscopy enables assessment of human melanoma microvasculature in vivo
Raster-Scanning-Optoacoustic Mesoscopy can be used to image the vasculature in skin cancer lesions but is limited by a long exposure time. Here; the authors increase the speed of the imaging using co-axial illumination and a high-sensitivity ultrasound detector path.
- Hailong He
- , Christine Schönmann
- & Vasilis Ntziachristos
-
Article
| Open AccessTargeted detection of cancer at the cellular level during biopsy by near-infrared confocal laser endomicroscopy
Radiography identifies suspicious lung nodules that are not always easy to diagnose via biopsy. Here, the authors utilize a fluorescent dye that targets the folate receptor and show using needle based endomicroscopy that it can be used to identify cancer cells during biopsy procedures
- Gregory T. Kennedy
- , Feredun S. Azari
- & Sunil Singhal