Review: Clinical in vivo multiphoton FLIM tomography
Methods Appl. Fluoresc. 8(2020)034002
Fluorescence Lifetime Imaging (FLIM) in life sciences based on ultrashort laser scanning microscopy and time-correlated single photon counting (TCSPC) started 30 years ago in Jena/East-Germany. One decade later, ﬁrst two-photon FLIM images of a human ﬁnger were taken with a lab microscope based on a tunable femtosecond Ti:sapphire laser. In 2002/2003, ﬁrst clinical non-invasive two-photon FLIM studies on patients with dermatological disorders were performed using a novel multiphoton tomograph. Current in vivo two-photon FLIM studies on human subjects are based on TCSPC and focus on (i) patients with skin inﬂammation and skin cancer as well as brain tumors, (ii) cosmetic research on volunteers to evaluate anti-ageing cremes, (iii) pharmaceutical research on volunteers to gain information on in situ pharmacokinetics, and (iv) space medicine to study non-invasively skin modiﬁcations on astronauts during long-term space ﬂights. Two-photon FLIM studies on volunteers and patients are performed with multiphoton FLIM tomographs using near infrared femtosecond laser technology that provide rapid non-invasive and label-free intratissue autoﬂuorescence biopsies with picosecond temporal resolution.
Translation of two-photon microscopy to the clinic: multimodal multiphoton CARS tomography of in vivo human skin
König K., Breunig HG., Batista A., Schindele A., Zieger M. Kaatz M.
J. Biomed. Opt. 25(1) (2020) 014515 (DOI: 10.1117/1.JBO.25.1.014515)
Two-photon microscopes have been successfully translated into clinical imaging tools to obtain high-resolution optical biopsies for in vivo histology. We report on clinical multi-photon coherent anti-Stokes Raman spectroscopy (CARS) tomography based on two tunable ultrashort near-infrared laser beams for label-free in vivo multimodal skin imaging. The multi-photon biopsies were obtained with the compact tomograph “MPTflex-CARS” using a photonic crystal fiber, an optomechanical articulated arm, and a four-detector-360 deg measurement head. The multiphoton tomograph has been employed to patients in a hospital with diseased skin. The clinical study involved 16 subjects, 8 patients with atopic dermatitis, 4 patients with pso-riasis vulgaris, and 4 volunteers served as control. Two-photon cellular autofluorescence life-time, second harmonic generation (SHG) of collagen, and CARS of intratissue lipids/proteins have been detected with single-photon sensitivity, submicron spatial resolution, and picosecond temporal resolution. The most important signal was the autofluorescence from nicotinamide adenine dinucleotide [NAD(P)H]. The SHG signal from collagen was mainly used to detect the epidermal–dermal junction and to calculate the ratio elastin/collagen. The CARS/Raman signal provided add-on information. Based on this view on the disease-affected skin on a sub-cellular level, skin areas affected by dermatitis and by psoriasis could be clearly identified. Multimodal multiphoton tomographs may become important label-free clinical high-resolution imaging tools for in vivo skin histology to realize rapid early diagnosis as well as treatment control.
Artificial Intelligence in Multiphoton Tomography: Atopic Dermatitis Diagnosis.
Guimaraes P., Batista A., Zieger M., Kaatz M., König K.
Scientific Reports 10(2020)7968
The diagnostic possibilities of multiphoton tomography (MPT) in dermatology have already been demonstrated. Nevertheless, the analysis of MPT data is still time-consuming and operator dependent. We propose a fully automated approach on convolutional neural networks (CNNs) to fully realize the potential of MPT. In total, 3,663 MPT images combining both morphological and metabolic information were acquired from atopic dermatitis (AD) patients and healthy volunteers. These were used to train and tune CNNs to detect the prescence of living cells, and if so, to diagnose AD, independently of imaged layer or position. The proposed algorithm correctly diagnosed AD in 97.00.2% of all images presenting living cells. The diagnosis was obtained with a sensitivity of 0.9660.003, specificity of 0.9770.003 and F-score of 0.9640.002. Relevance propagation by deep Taylor decomposition was used to enhance the algorithm’s interpretability. Obtained heatmaps show what aspects of the images are important for a given classification. We showed that MPT imaging can be combined with artificial intelligence to successfully diagnose AD. The proposed approach serves as a framework for the automatic diagnosis of skin disorders using MPT.
Non-invasive optical biopsy by multiphoton microscopy identifies the live morphology of common melanocytic nevi
Lentsch G, Valdebran M, Saknite I, Smith J, Linden KG, König K, Barr RJ, | Harris RM, Tromberg BJ, Ganesan AK, Zachary CB, Kelly KM, Balu M.
Pigment Cell & Melanoma Res. (2020)001-9 (DOI 10.1111/pcmr.12902)
Multiphoton microscopy (MPM) is a promising non-invasive imaging tool for discrimi-nating benign nevi from melanoma. In this study, we establish a MPM morphologic catalogue of common nevi, information that will be critical in devising strategies to distinguish them from nevi that are evolving to melanoma that may present with more subtle signs of malignancy. Thirty common melanocytic nevi were imaged in vivo using MPM. Quantitative parameters that can distinguish between different types of nevi were developed and confirmed by examining the histology of eleven of the imaged nevi. MPM features of nevi examined included cytologic morphology of melanocytes in the epidermis and dermis, the size and distribution of nevomelano-cytes both within and around nests, the size of rete ridges, and the presence of im-mune cells in the dermis. Distinguishing features include cytological morphology, the size of nevomelanocytes, the size of nevomelanocyte nests, and the distribution of nevomelanocytes. Notably, these distinguishing characteristics were not easily ap-preciated in fixed tissues, highlighting essential differences in the morphology of live skin. Taken together, this work provides a morphologic compendium of normal nevi, information that will be critical in future studies directed at identifying melanocytic nevi that are evolving to melanoma.