Skip to content


High-Resolution Raman Imaging of >300 Patient-Derived Cells from Nine Different Leukemia Subtypes: A Global Clustering Approach

Renzo Vanna, Andrea Masella, Manuela Bazzarelli, Paola Ronchi, Aufried Lenferink, Cristina Tresoldi, Carlo Morasso, Marzia Bedoni, Giulio Cerullo, Dario Polli, Fabio Ciceri, Giulia De Poli, Matteo Bregonzio, Cees Otto.

Analytical Chemistry, 2024

DOI: 10.1021/acs.analchem.4c00787

Leukemia comprises a diverse group of bone marrow tumors marked by cell proliferation. Current diagnosis involves identifying leukemia subtypes through visual assessment of blood and bone marrow smears, a subjective and time-consuming method. Our study introduces the characterization of different leukemia subtypes using a global clustering approach of Raman hyperspectral maps of cells. We analyzed bone marrow samples from 19 patients, each presenting one of nine distinct leukemia subtypes, by conducting high spatial resolution Raman imaging on 319 cells, generating over 1.3 million spectra in total. An automated preprocessing pipeline followed by a single-step global clustering approach performed over the entire data set identified relevant cellular components (cytoplasm, nucleus, carotenoids, myeloperoxidase (MPO), and hemoglobin (HB)) enabling the unsupervised creation of high-quality pseudostained images at the single-cell level. Furthermore, this approach provided a semiquantitative analysis of cellular component distribution, and multivariate analysis of clustering results revealed the potential of Raman imaging in leukemia research, highlighting both advantages and challenges associated with global clustering.

Intraoperative Assessment of Tumor Margins in Tissue Sections with Hyperspectral Imaging and Machine Learning

David Pertzborn, Hoang-Ngan Nguyen, Katharina Hüttmann, Jonas Prengel, Günther Ernst, Orlando Guntinas-Lichius, Ferdinand von Eggeling, Franziska Hoffmann

Cancers, 2023

DOI: 10.3390/cancers15010213

The complete resection of the malignant tumor during surgery is crucial for the patient’s survival. To date, surgeons have been intraoperatively supported by information from a pathologist, who performs a frozen section analysis of resected tissue. This tumor margin evaluation is subjective, methodologically limited and underlies a selection bias. Hyperspectral imaging (HSI) is an established and rapid supporting technique. New artificial-intelligence-based techniques such as machine learning (ML) can harness this complex spectral information for the verification of cancer tissue. We performed HSI on 23 unstained tissue sections from seven patients with oral squamous cell carcinoma and trained the ML model for tumor recognition resulting in an accuracy of 0.76, a specificity of 0.89 and a sensitivity of 0.48. The results were in accordance with the histopathological annotations and do, therefore, enable the delineation of tumor margins with high speed and accuracy during surgery.

A novel interpretable machine learning algorithm to identify optimal parameter space for cancer growth

Coggan Helena, Andres Terre Helena, Liò Pietro.

Frontiers in Big Data, 2022

DOI: 10.3389/fdata.2022.941451

Recent years have seen an increase in the application of machine learning to the analysis of physical and biological systems, including cancer progression. A fundamental downside to these tools is that their complexity and nonlinearity makes it almost impossible to establish a deterministic, a priori relationship between their input and output, and thus their predictions are not wholly accountable. We begin with a series of proofs establishing that this holds even for the simplest possible model of a neural network; the effects of specific loss functions are explored more fully in Appendices. We return to first principles and consider how to construct a physics-inspired model of tumor growth without resorting to stochastic gradient descent or artificial nonlinearities. We derive an algorithm which explores the space of possible parameters in a model of tumor growth and identifies candidate equations much faster than a simulated annealing approach. We test this algorithm on synthetic tumor-growth trajectories and show that it can efficiently and reliably narrow down the area of parameter space where the correct values are located. This approach has the potential to greatly improve the speed and reliability with which patient-specific models of cancer growth can be identified in a clinical setting.