Design and clinical validation of a point-of-care device for the diagnosis of lymphoma via contrast-enhanced microholography and machine learning

Abstract

The identification of patients with aggressive cancer who require immediate therapy is a health challenge in low-income and middle-income countries. Limited pathology resources, high healthcare costs and large-case loads call for the development of advanced standalone diagnostics. Here, we report and validate an automated, low-cost point-of-care device for the molecular diagnosis of aggressive lymphomas. The device uses contrast-enhanced microholography and a deep-learning algorithm to directly analyse percutaneously obtained fine-needle aspirates. We show the feasibility and high accuracy of the device in cells, as well as the prospective validation of the results in 40 patients clinically referred for image-guided aspiration of nodal mass lesions suspicious for lymphoma. Automated analysis of human samples with the portable device should allow for the accurate classification of patients with benign and malignant adenopathy.

Use of artificial intelligence in infectious diseases

“…Even in developing countries where the resources, health-care cost, and other limitations prevent from providing optimal care, this is applicable. A group has recently shown the possibility to develop a low-cost point of care for lymphoma diagnosis based on basic imaging and deep learning (Im et al, 2018). Several reports suggested the use of Bayesian network (BN) for representing statistical dependencies (Xu et al, 2016).…”

Non‐viral gene delivery for cancer immunotherapy

“…macrophage and DCs) in the lymph nodes, the emerging strategies of cancer immunotherapy by actively modulating the APCs in the TME have also attracted extensive attention . Im et al . developed TLR7/8 agonist‐loaded β‐cyclodextrin NPs (CDNP‐R848) for efficient drug delivery to tumour‐associated macrophages in vivo .…”

“…The administration of CDNP‐R848 altered the functional orientation of the TAM towards an M1 phenotype, leading to growth inhibition in a mouse model of M38 colorectal tumor and B16‐F10 melanoma tumors. The antitumor efficacy of the CDNP‐R848 was further improved by combination with the immune checkpoint inhibitor anti‐PD‐1, implying the potential of drug/gene delivery to the TAM‐associated macrophages for cancer immunotherapy . However, challenges including weak immunogenicity, systemic toxicity and off‐target effects of cancer vaccines hinder their broad clinical translation.…”

“…macrophage and DCs) in the lymph nodes, the emerging strategies of cancer immunotherapy by actively modulating the APCs in the TME have also attracted extensive attention. 96 Im et al 97…”

Spectroscopy and Machine Learning Based Rapid Point-of-Care Assessment of Core Needle Cancer Biopsies

“…9 In pursuing cancer biopsy quality improvement, a variety of technologies and techniques have been considered for point-of-care in vivo or ex vivo tissue characterization. 10 These include optical spectroscopy, 11 x-ray imaging, 12 confocal microscopy, [13][14][15] structured illumination microscopy, 16 Fourier transform infrared imaging, 17 fluorescence microscopy, [18][19][20][21] contrast-enhanced micrography, 22 as well as diffuse reflectance, electrical impedance, and Raman spectroscopy. [23][24][25][26] Use of these technologies in a clinical setting has been hampered by factors such as lengthy analytic times, tissue degradation, expense, on-site tissue staining, requirement for interpretive expertise, and challenges to workflow integration.…”