Classification of breast cancer from histopathology images using an ensemble of deep multiscale networks

Journal of Applied Biomedicine, Год журнала: 2022, Номер 42(3), С. 963 - 976

Опубликована: Июль 1, 2022

Язык: Английский

---

Perspective on quantitative phase imaging to improve precision cancer medicine

Journal of Biomedical Optics, Год журнала: 2024, Номер 29(S2)

Опубликована: Март 26, 2024

SignificanceQuantitative phase imaging (QPI) offers a label-free approach to non-invasively characterize cellular processes by exploiting their refractive index based intrinsic contrast. QPI captures this contrast translating associated shifts into intensity-based quantifiable data with nanoscale sensitivity. It holds significant potential for advancing precision cancer medicine providing quantitative characterization of the biophysical properties cells and tissue in natural states.AimThis perspective aims discuss increase our understanding development its response therapeutics. also explores new developments methods towards personalized therapy early detection.ApproachWe begin detailing technical advancements QPI, examining implementations across transmission reflection geometries retrieval methods, both interferometric non-interferometric. The focus then QPI's applications research, including dynamic cell mass drug assessment, risk stratification, in-vivo imaging.ResultsQPI has emerged as crucial tool medicine, offering insights tumor biology treatment efficacy. Its sensitivity detecting changes promise enhancing diagnostics, prognostication. future is envisioned integration artificial intelligence, morpho-dynamics, spatial biology, broadening impact research.ConclusionsQPI presents redefining diagnosis, monitoring, treatment. Future directions include harnessing high-throughput imaging, 3D realistic models, combining intelligence multi-omics extend capabilities. As result, stands at forefront research clinical application care.

Язык: Английский

---

Label-free cell classification in holographic flow cytometry through an unbiased learning strategy

Lab on a Chip, Год журнала: 2024, Номер 24(4), С. 924 - 932

Опубликована: Янв. 1, 2024

Unbiased learning pipeline for label-free single-cell classification.

Язык: Английский

---

Phenotyping neuroblastoma cells through intelligent scrutiny of stain-free biomarkers in holographic flow cytometry

APL Bioengineering, Год журнала: 2023, Номер 7(3)

Опубликована: Сен. 1, 2023

To efficiently tackle certain tumor types, finding new biomarkers for rapid and complete phenotyping of cancer cells is highly demanded. This especially the case most common pediatric solid sympathetic nervous system, namely, neuroblastoma (NB). Liquid biopsy in principle a very promising tool this purpose, but usually enrichment isolation circulating such patients remain difficult due to unavailability universal NB cell-specific surface markers. Here, we show that screening through stain-free supported by artificial intelligence viable route liquid biopsy. We demonstrate concept flow cytometry based on label-free holographic quantitative phase-contrast microscopy empowered machine learning. In detail, exploit hierarchical decision scheme where at first level are classified from monocytes with 97.9% accuracy. Then different phenotypes discriminated within class. Indeed, each cell as its belonging one four sub-populations (i.e., CHP212, SKNBE2, SHSY5Y, SKNSH) evaluated thus achieving accuracy range 73.6%-89.1%. The achieved results solve realistic problem related identification cell, i.e., possibility recognize detect morphologically similar blood cells, which core issue microscopy. presented approach operates lab-on-chip scale emulates real-world scenarios, representing future exploiting intelligent biomedical imaging.

Язык: Английский

---

On the hydrodynamic mutual interactions among cells for high-throughput microfluidic holographic cyto-tomography

Optics and Lasers in Engineering, Год журнала: 2022, Номер 158, С. 107190 - 107190

Опубликована: Июль 22, 2022

In digital holography (DH) modality for lab on chip application the cells passing through field of view (FOV) microscope can be detected and analyzed even if they are flowing at different depths. fact, in-focus imaging each cell easily retrieved thanks to ability DH obtain numerical focus ex-post recording process. An advantageous preferred configuration in flow-cytometry provides that rotate along microfluidic channel. This gives unique chance probing by light beams alongside many directions while cross holographic FOV. Thus, it is possible retrieve 3D refractive index map cell, i.e. a phase-contrast tomogram. Although same FOV, thus giving possibility increase throughput system, until now no investigations have been made establish how close avoid mutual disturbing effects their rotation. Nevertheless, estimate maximum achievable throughput, indispensable comprehend hydrodynamic interactions adjacent tomographic flow cytometer. Here we show means an experimental simulation fluid dynamics quantitative effect rotational behavior neither mechanical deformation. However, considered scenario demonstrate negligible as does not affect recovering tomograms. The reported results will allow which optimum density analyze flow-cyto-tomograph opening route biomedical applications.

Язык: Английский

---

Multiscale optical phase fluctuations link disorder strength and fractal dimension of cell structure

Biophysical Journal, Год журнала: 2023, Номер 122(7), С. 1390 - 1399

Опубликована: Март 5, 2023

Язык: Английский

---

Holography - Recent Advances and Applications

IntechOpen eBooks, Год журнала: 2022, Номер unknown

Опубликована: Июль 22, 2022

Holography of today is a broad field developed in the meeting between optics and digital world computers. A hologram usually contains more or different information on observed scene than regular image same scene. The development has been accelerated lately due to improvement cameras, computers, light sources, spatial modulators. As multidisciplinary area, holography connects experts electro-optical engineering, processing, computer algorithms. More are needed when utilized various applications such as microscopy, industrial inspection, biomedicine, entertainment. This book provides an overview from aspect concepts, system architectures, applications.

Язык: Английский

---

Automated Stain‐Free Holographic Image‐Based Phenotypic Classification of Elliptical Cancer Cells

Advanced Photonics Research, Год журнала: 2022, Номер 4(1)

Опубликована: Окт. 17, 2022

Image‐based stain‐free elliptical cancer cell classification is very challenging due to interclass morphological similarity. Herein, the of three types lines (lung, breast, and skin) by feature‐based machine learning image‐based deep with a convolutional neural network (CNN) addressed. Digital holography in microscopic configuration used obtain quantitative phase images representing intracellular content morphology cells. In classification, several features related both material thickness cells are extracted, followed feature selection training random forest, support vector machine, pattern recognition artificial networks. For two CNN models trained: skip connections (Resnet) without connection. The accuracy strategies analyzed strategy outperforms about 9% 10‐fold cross‐validation evaluation.

Язык: Английский

---

Multimodal segmentation of dynamic subcellular features using quantitative phase imaging and FRET-based sensors [Invited]

Journal of the Optical Society of America A, Год журнала: 2024, Номер 41(11), С. C38 - C38

Опубликована: Сен. 12, 2024

Understanding cellular responses to mechanical environmental stimuli is important for mechanotransduction studies. While fluorescence microscopy has been used aiding research due its molecular sensitivity, the ability of quantitative phase imaging (QPI) visualize morphology yet be widely applied, perhaps limited specificity. Here, we seek expand on previous work which combined with a molecularly sensitive Förster resonance energy transfer (FRET) construct by developing additional analysis techniques. This seeks characterize response individual cells stimulus through novel, best our knowledge, QPI-guided segmentation algorithm. The multimodal instrument and techniques are employed examine hypo-osmotic observing calcium ion flux using FRET-based sensor coupled mapping intracellular mass reorganization QPI. modality enables discrimination cell localized region, revealing distinct behavior between regions relative control group. Our novel can identify expansion region specific in both modalities stimulus. With broad array FRET sensors under development, complementary addition QPI offers new avenues studying range stimuli.

Язык: Английский

---

Biophysical Profiling of Sickle Cell Disease Using Holographic Cytometry and Deep Learning

International Journal of Molecular Sciences, Год журнала: 2023, Номер 24(15), С. 11885 - 11885

Опубликована: Июль 25, 2023

Sickle cell disease (SCD) is an inherited hematological disorder associated with high mortality rates, particularly in sub-Saharan Africa. SCD arises due to the polymerization of sickle hemoglobin, which reduces flexibility red blood cells (RBCs), causing vessel occlusion and leading severe morbidity early rates if untreated. While solubility tests are available African population as a means for detecting hemoglobin (HbS), test falls short assessing severity visualizing degree cellular deformation. Here, we propose use holographic cytometry (HC), throughput, label-free imaging modality, comprehensive morphological profiling RBCs detect SCD. For this study, more than 2.5 million single-cell images from normal patient samples were collected using HC system. We have developed approach specially defining training data improve machine learning classification. demonstrate deep classifier can produce highly accurate classification, even on unknown samples.

Язык: Английский

---