Review of Fluorescence Lifetime Imaging Microscopy (FLIM) Data Analysis Using Machine Learning

Adhikari, Mou; Houhou, Rola; Hniopek, Julian; Bocklitz, Thomas

doi:10.3390/jeta1010004

Cited by 9 publications

(2 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An added advantage of the dual emissive on-to-on fluorophore design is that the dynamic interrelationship between both states can also be scrutinized through their different lifetimes using FLIM and phasor plot data analysis [ 49 ]. Phasor plot analysis allows for the graphical representation of the raw fluorescence lifetime data, such that each pixel in a FLIM image is transformed to a point in the phasor plot, with pixels containing a combination of two different lifetimes graphed according to the weighted linear combination of their contributions [ 50 ].…”

Section: Resultsmentioning

confidence: 99%

A NIR-Fluorochrome for Live Cell Dual Emission and Lifetime Tracking from the First Plasma Membrane Interaction to Subcellular and Extracellular Locales

Booth,

Garre,

et al. 2024

Molecules

View full text Add to dashboard Cite

Molecular probes with the ability to differentiate between subcellular variations in acidity levels remain important for the investigation of dynamic cellular processes and functions. In this context, a series of cyclic peptide and PEG bio-conjugated dual near-infrared emissive BF2-azadipyrromethene fluorophores with maxima emissions at 720 nm (at pH > 6) and 790 nm (at pH < 5) have been developed and their aqueous solution photophysical properties determined. Their inter-converting emissions and fluorescence lifetime characteristics were exploited to track their spatial and temporal progression from first contact with the plasma membrane to subcellular locales to their release within extracellular vesicles. A pH-dependent reversible phenolate/phenol interconversion on the fluorophore controlled the dynamic changes in dual emission responses and corresponding lifetime changes. Live-cell confocal microscopy experiments in the metastatic breast cancer cell line MDA-MB-231 confirmed the usability of the dual emissive properties for imaging over prolonged periods. All three derivatives performed as probes capable of real-time continuous imaging of fundamental cellular processes such as plasma membrane interaction, tracking endocytosis, lysosomal/large acidic vesicle accumulation, and efflux within extracellular vesicles without perturbing cellular function. Furthermore, fluorescence lifetime imaging microscopy provided valuable insights regarding fluorophore progression through intracellular microenvironments over time. Overall, the unique photophysical properties of these fluorophores show excellent potential for their use as information-rich probes.

show abstract

Section: Resultsmentioning

confidence: 99%

A NIR-Fluorochrome for Live Cell Dual Emission and Lifetime Tracking from the First Plasma Membrane Interaction to Subcellular and Extracellular Locales

Booth,

Garre,

et al. 2024

Molecules

View full text Add to dashboard Cite

show abstract

“…The explosive development of deep learning (DL) technologies is transforming conventional biomedical imaging analysis 12 , 13 , leading to automated processing that could surpass human capabilities. One of the beneficial fields is virtual histological staining from microscopic images acquired by various imaging modalities 14 using DL models, among which U-Net 15 -based Generative Adversarial Networks (GANs) 16 are prominent.…”

Section: Introductionmentioning

confidence: 99%

Deep learning-based virtual H& E staining from label-free autofluorescence lifetime images

Wang,

Akram,

Dorward

et al. 2024

npj Imaging

View full text Add to dashboard Cite

Label-free autofluorescence lifetime is a unique feature of the inherent fluorescence signals emitted by natural fluorophores in biological samples. Fluorescence lifetime imaging microscopy (FLIM) can capture these signals enabling comprehensive analyses of biological samples. Despite the fundamental importance and wide application of FLIM in biomedical and clinical sciences, existing methods for analysing FLIM images often struggle to provide rapid and precise interpretations without reliable references, such as histology images, which are usually unavailable alongside FLIM images. To address this issue, we propose a deep learning (DL)-based approach for generating virtual Hematoxylin and Eosin (H&E) staining. By combining an advanced DL model with a contemporary image quality metric, we can generate clinical-grade virtual H&E-stained images from label-free FLIM images acquired on unstained tissue samples. Our experiments also show that the inclusion of lifetime information, an extra dimension beyond intensity, results in more accurate reconstructions of virtual staining when compared to using intensity-only images. This advancement allows for the instant and accurate interpretation of FLIM images at the cellular level without the complexities associated with co-registering FLIM and histology images. Consequently, we are able to identify distinct lifetime signatures of seven different cell types commonly found in the tumour microenvironment, opening up new opportunities towards biomarker-free tissue histology using FLIM across multiple cancer types.

show abstract

The Quantitative In Vivo Assessment of Diabetic and Non‐Diabetic Skin Wound Healing Using Phasor‐FLIM Approach

Zuhayri,

Lepekhina,

Nikolaev

et al. 2024

Journal of Biophotonics

View full text Add to dashboard Cite

A quantitative assessment of wound status in a murine model was developed using phasor plot presentation of fluorescence lifetime imaging microscopy (FLIM) data. The quantitative assessment is based on calculating Bhattacharyya distance between g coordinates of FLIM data phasor plot density distributions of wound and healthy skin. The approach was validated for both diabetic and non‐diabetic mice wounds, including during low‐dose photodynamic therapy (LDPDT). Analysis revealed a shift in the FLIM data phasor plot g coordinates, suggesting altered metabolic processes involved in wound healing. Bhattacharyya distances in the LDPDT groups were closer to zero compared to the control group, which was not treated by LDPDT. Bhattacharyya distances in the non‐diabetic LDPDT groups were closer to zero compared to the diabetic LDPDT groups that is consistent with the literature regarding the positive role of LDPDT in accelerating wound healing and the role of diabetes mellitus in impairing wound healing.

show abstract

Review of Fluorescence Lifetime Imaging Microscopy (FLIM) Data Analysis Using Machine Learning

Cited by 9 publications

References 60 publications

A NIR-Fluorochrome for Live Cell Dual Emission and Lifetime Tracking from the First Plasma Membrane Interaction to Subcellular and Extracellular Locales

A NIR-Fluorochrome for Live Cell Dual Emission and Lifetime Tracking from the First Plasma Membrane Interaction to Subcellular and Extracellular Locales

Deep learning-based virtual H& E staining from label-free autofluorescence lifetime images

The Quantitative In Vivo Assessment of Diabetic and Non‐Diabetic Skin Wound Healing Using Phasor‐FLIM Approach

Contact Info

Product

Resources

About