Early detection of Alzheimer's
Alzheimer's disease (AD) is the most common form of dementia, resulting in cognitive impairments that affect memory, thinking, and behavior. The disease affects an estimated 44 million people worldwide in an increasingly aging population with higher life expectancy. Since there are currently no treatments available to cure the disease, only to alleviate symptoms, research is focused on developing new drugs to delay the onset of the disease. In this sense, early detection of the disease has become one of the most important means to combat the disease and minimize its negative effects.
Biomarkers of the disease play an important role in early non-invasive diagnosis of biofluids because in the earliest stages of the disease, amyloid-beta proteins (Aβ) and both total and phosphorylated tau proteins (t-tau and -tau) become abnormal in cerebrospinal fluid (CSF) and blood, even before the disease is detected by positron emission tomography (PET) and before neurodegeneration appears with symptoms. However, ongoing research shows that there are many other components of CSF that may reflect the development of AD, such as phospholipids, fatty acids, or apolipoproteins. Therefore, considering the entirety of biochemical compunds in body fluids will help develop more effective and reliable techniques for the early detection of AD.
Photonic techniques have the unique ability to study biosamples in real-time without destroying the sample. Vibrational spectroscopic techniques have the advantage of high chemical specificity, which makes them ideal for obtaining a broad biochemical spectrum of the biosamples analyzed.
The goal of this project is to develop holistic analytical tools for in vitro diagnostics supported by machine learning. In general, non-invasive samples such as blood or plasma from both patients and healthy controls are analyzed by image analysis on the one hand and spectroscopic techniques such as Raman, surface-enhanced Raman (SERS), or Fourier-transform infrared (FTIR) on the other. In this way, we obtain comprehensive information about the biochemical composition of the samples, which is later analyzed and combined using machine learning to obtain a holistic picture of the patient's health status with regard to the risk of developing AD. This project is carried out by a 4-member consortium, consisting of CIC nanoGUNE as organizer, CITA-Alzheimer, Vicomtech and the University of the Basque Country UPV/EHU (IBeA).
L. A. Arévalo, S. A. O’Brien, O. Antonova, and A. Seifert, “Drying patterns of cerebrospinal fluid as indicator for Alzheimer’s disease by a machine learning framework,” Journal of Physics: Conference Series, vol. 2407, no. 1, p. 012027, 2022.
L. A. Arévalo, O. Antonova, S. A. O’Brien, G. P. Singh, and A. Seifert, “Detection of Alzheimer’s by machine learning-assisted vibrational spectroscopy in human cerebrospinal fluid,” Journal of Physics: Conference Series, vol. 2407, no. 1, p. 012026, 2022.