A study finds image analysis using machine learning can identify haematological malignancies.

Image analysis is typically used to extract meaningful information from images. It can perform tasks like finding shapes, identifying edges, removing noise, counting objects, etc. for image quality. Recently, a study demonstrated that image analysis utilizing neural networks can help detect details in tissue samples that are difficult to determine with the human eye. Myelodysplastic syndrome (MDS) is a disease of the stem cells in the bone marrow, which affects the maturation and differentiation of blood cells. Diagnosing MDS requires a bone marrow sample to investigate genetic changes in the bone marrow cells.

Annually, some 200 Finns are diagnosed with MDS, which can develop into acute leukaemia. The incidence of MDS globally is 4 cases per 100,000 person years. The syndrome is classified into groups to find out the nature of the disorder in more detail.

In the University of Helsinki study, microscopic images of patients' bone marrow samples suffering from myelodysplastic syndrome were analysed utilising an image analysis technique based on machine learning. The samples were stained with haematoxylin and eosin (H&E staining), a procedure of routine diagnostics for the disease. The slides were digitised and analysed using computational deep learning models.

The study was published in the Blood Cancer Discovery, a journal of the American Association for Cancer Research. The results can be explored with an interactive tool: http://hruh-20.it.helsinki.fi/mds_visualization/.

With machine learning, the digital image dataset could be assessed to accurately identify the most common genetic mutations affecting the progression of the syndrome, such as acquired mutations and chromosomal aberrations. The higher the number of abnormal cells in the samples, the higher the reliability of the results generated by the prognostic models.

The study uses the data analysis technique to support the diagnosis. One of the greatest challenges of leveraging neural network models is to understand the criteria on which they base their conclusions drawn from data, such as information contained in images. The University of Helsinki study succeeded in determining what deep learning models see in tissue samples when they have been taught to look for, for example, genetic mutations related to MDS. The technique provides new information on the effects of complex diseases on bone marrow cells and the surrounding tissues.

According to Professor Satu Mustjoki, 'the study confirms that computational analysis helps to identify features that elude the human eye. Moreover, data analysis helps to collect quantitative data on cellular changes and their relevance to the patient's prognosis.'

Part of the analytics carried out in the study was implemented using the Helsinki University Hospital (HUS) data lake environment, which enables the efficient collection and analysis of extensive clinical datasets.

"We've developed solutions to structure and analyse data stored in the HUS data lake. Image analysis helps us analyse large quantities of biopsies and rapidly produce diverse information on disease progression. The techniques developed in the project are suited to other projects as well, and they are perfect examples of digitalizing medical science," says doctoral student Oscar Bruck.

Ph.D. Olivier Elemento from the Caryl and Israel Englander Institute for Precision Medicine says, "[This] study provides new insights into the pathobiology of MDS and paves the way for increased use of artificial intelligence for the assessment and diagnosis of hematological malignancies."