Knowing the results of root canal therapy is crucial to support clinical decision-making on whether the tooth will be retained in the mouth. Both clinically and radiographically, the outcome of root canal therapy can be assessed subjectively. Several studies have reported the uncertainty of the degree of association between histological findings and radiographic appearance of periapical tissues. Therefore, clinical characteristics and the radiographic reviews are the most practical methods to specify the degree of healing after root canal treatment [1,2]. Ørstavik et al. suggested using the “periapical index” (PAI) when interpreting the results of root canal treatment radiographically [3].

This developing lesion can be referred to as failed endodontic treatment or post-endodontic disease (post-treatment disease). The causes of fail consist of intrinsic or extrinsic nonmicrobial causes as well as microbial factors, such as extraradicular and/or intraradicular infections [4]. In primary endodontic infections and associated symptomatic and asymptomatic endodontic infections, anaerobic gram-negative bacteria are the most common microorganisms that disrupt the sterilization of the root-canal system that cause abscess associated with apical periodontitis [[5], [6], [7]]. The most common extraradicular infection due to intraradicular infection is an acute apical abscess. Extraradicular microorganisms have been implicated as one of the causes of persistent apical periodontitis lesions that occur despite proper root-canal treatment [8]. In clinical practice, it is possible to classify different types of apical periodontitis using the PAI, which tracks the severity of lesions [9,10]. Therefore, it is of great importance to identify endodontic bacteria with high-throughput techniques, which may explain the severity of lesions identified by the PAI scores or be responsible for determining molecular mechanisms [11].

Microorganisms that cannot be removed in the apical 1/3 of the root canal and persist in the cementum lacunae close to the apical foramen are among the causes of failure of root canal treatments [12]. Specific bacteria associated with painful exacerbations, periapical destruction, and persistent infections in endodontics and the identification of these bacteria are extremely important and current issues [8]. The necessity of identifying the pathogens that cause endodontic diseases is important in increasing the success rate of root canal treatments.

Microscopy, culture, immunological, and molecular biological methods are used to detect root canal pathogens. In studies, we see that polymerase chain reaction (PCR) is the most used method among these methods. However, it has many disadvantages, such as being a laborious and expensive method and detecting dead cells along with living cells. It has been stated that various methods in molecular biology respond in a short time for the detection of these microorganisms [13]. Advances in molecular biology have focused on matrix assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS). Through the soft ionization approach of MALDI-TOF MS and the desorption of bacterial proteins into ions, specific mass spectra are identified, and the identity of the bacteria is determined within minutes [14]. It is one of the cheaper, faster and more accurate detection methods compared to molecular and immunological based methods [[15], [16], [17], [18], [19]]. Because of such advantages, developments in molecular biology have focused on MALDI-TOF MS. Widely used for protein analysis, MALDI-TOF MS is a potential tool for microbial diagnosis [20,21].

The purpose of this study was to use MALDI-TOF-MS to identify bacteria isolated from various periapical tissue diseases and classify them using an unsupervised machine learning approach.