How artificial intelligence makes life easier for dentists and patients
Artificial intelligence has become one of the hottest topics over the past few years. ChatGPT writes books and theses, DALL-E creates paintings based on text descriptions, streaming services such as Kinopoisk use artificial intelligence to create recommendations, voice assistants answer almost any user request. Medicine and dentistry are no exception – artificial intelligence is changing the practice of doctors. According to a study published in Journal of Dental Sciencesthanks to the integration of artificial intelligence technology and digital imaging systems, dentists can now quickly analyze X-ray and CT images and identify oral health problems such as tooth decay, bone loss or infection.
Some doctors and patients are wary of the use of algorithms in medicine. But it is important to understand that algorithms do not replace physician diagnosis: in the clinical process, artificial intelligence is simply another tool that helps doctors make diagnoses more consistently and effectively. Artificial intelligence can be compared to a microscope: it also helps to see what is difficult to see with the naked eye, and gives dentists more data for diagnosis and treatment. Both artificial intelligence and a microscope improve the clinical skills of a doctor, but do not replace the specialist himself.
Let's look at the example of Diagnocat, an X-ray image analysis service based on artificial intelligence, how algorithms help automate labor-intensive tasks, improve the quality and comfort of patient care, and reduce diagnostic time.
What is Diagnocat?
This is a program that, based on 2D and 3D images, generates convenient reports for doctors on the condition of the dental system.
Each tooth on an x-ray is automatically identified using the Universal Dental Tooth Numbering System, and a report is then generated that describes the condition of each tooth. Diagnocat can identify 20 possible indicators for 2D images and over 65 for CBCT. The program can also detect missing teeth. In addition to the program annotations, the doctor can add his own comments to the report for each tooth. The company positions Diagnocat as a reliable virtual assistant: it simply generates [1] a detailed report based on radiographs, and the doctor decides what to do with this information.
Diagnocat allows you to collect, differentiate and group a large amount of information in less than 10 minutes. Therefore, at the Confidence clinic, the results of diagnostic examinations of each primary patient are passed through the prism of artificial intelligence. An effective and comfortable system is built as follows:
1) The patient arrives 20-30 minutes before the appointment and undergoes a computed tomography scan of the dental system
2) The X-ray technician uploads the image into the Diagnocat program.
3) By the beginning of the consultation, he receives a photo with artificial intelligence annotations. This is an almost complete package of information that is necessary to draw up a treatment plan: all that is missing is photographs and a discussion of the patient’s complaints. For example, for experienced orthodontists, information from a CBCT image and analysis from Diagnocat, photographs taken at an appointment, is enough to formulate a final treatment plan.
Fast, accurate, unbiased
News about artificial intelligence achievements appears every month, so there is no doubt that artificial intelligence is capable of many things. The main advantages of using Diagnocat in medicine are the speed of image processing, reduction of diagnostic errors, increased trust in the patient-doctor relationship and cooperation between doctors.
1) Processing speed
To diagnose each patient, doctors take from 30 minutes to 1.5 hours. During this time, the doctor analyzes and structures all the information based on a number of examinations: 3D cephalometric analysis, orthopantomogram, dental photographs, analysis of the temporomandibular joint and respiratory tract. Generating the same report using Diagnocat takes less than 10 minutes. This speed of diagnosis significantly reduces the appointment time for the doctor and the waiting period for the patient.
2) Reducing the number of errors and missed problems
Artificial intelligence cannot miss sleep or be in a bad mood because of a missed cup of coffee in the morning. He cannot be tired or inattentive at the end of the working day. Artificial intelligence maintains its accuracy and performance at all times. In addition, Diagnocat distinguishes many more “shades of gray” in an X-ray image than the human eye can see. This avoids diagnostic inaccuracies when reading radiographs that have been overexposed or underexposed. At the same time, the program searches for up to 65 different images in a 3D image, often this is more than a dentist can keep in mind. All together, these advantages make it possible to avoid situations where some problems with the dental system escape the attention of the radiologist.
Unfortunately, this does not mean that Diagnocat recognizes all conditions, but the company that created it claims that screening accuracy is achieved for some conditions. The reports generated by the program can be used as a second independent opinion, but the diagnosis itself is made by the doctor.
3) Patient acceptance of the treatment plan
Unfortunately, sometimes the patient does not trust the doctor and expects the dentist to inflate the price of treatment or prescribe additional unnecessary procedures in order to “sell” more of his services. The artificial intelligence report in this case is considered by the patient as a neutral third party – after all, the algorithm cannot derive any benefit from the results. People sometimes trust the program more than the doctor. In addition, Diagnocat creates a visually beautiful picture for discussion that is understandable for the patient. Therefore, based on the generated report from the algorithm, it is usually easier for the doctor to explain to the patient the necessary treatment plan and convey the harmful consequences of refusing the planned plan.
4) Cooperation
Diagnocat prepares a report using a standardized form, and also allows you to view a report on a patient from anywhere in the world: just go through a browser, select a patient and view all the information. This facilitates the cooperation of doctors in cases where complex treatment is required: the orthodontist, therapist and surgeon can view information about the patient at any time and offer an optimal treatment strategy.
What are the benefits of using Diagnocat for an orthodontist?
An orthodontic report prepared in Diagnocat is as convenient as possible for diagnosing bite problems, discussing a plan with the patient and treatment. It includes:
Visually beautiful, clear picture for discussion with the patient
3D cephalometric analysis with comments
Teleradiogram on the right and left with an outline outlining the teeth and jaw
Slicing of each tooth, which saves time on viewing and searching for teeth in regular viewers
Sagittal sections of the TMJ
Coronal sections of the TMJ
Airway assessment
Screening of every tooth
It is very important that Diagnocat allows you to upload three-dimensional models of each tooth to any other programs, since the use of a complex of several programs allows you to further reduce the time of an appointment with an orthodontist. For example, STL 3D models of teeth can be uploaded to Maestro, a program that orthodontists use to simulate the movement of teeth on aligners or to create a tray for indirect braces. (Indirect fixation is a method in which braces are not installed one at a time, but are transferred directly to the dentition using plastic “transfer” trays. This method of fixation allows you to reduce the installation time of braces from 1-1.5 hours to 30-40 minutes).
The program also relieves the doctor of the need for monotonous actions. Before using an artificial intelligence-based program, in order to assess the position of each tooth in the bone tissue in our clinic’s experience, it was necessary to take screenshots of each tooth. This is a very monotonous task that takes about half an hour. Now that artificial intelligence does the cutting itself, we have a lot more time for analysis. Once the file is saved in Diagnocat, we can return to it at any time.
Diagnocat and dental therapy.
The program itself marks healthy teeth, teeth that have undergone treatment and problem teeth, visually highlights them in color and gives the percentage probability of a particular pathology. In addition to the available slices, the doctor can open the 3D Viewer and interact with the slices. The therapist can also change the brightness and contrast of the image if necessary and quickly print the resulting images to show the patient the abnormalities and suggest treatment.
The program can also prepare an endodontic report: it provides detailed sections of the root canal system, the furcation area (the process of bone loss in the area of branching of the tooth roots). AI automatically measures the volume of the periapical lesion, thereby allowing the doctor to choose the necessary tactics and assess the dynamics of treatment.
Artificial intelligence for the surgeon.
Surgeons can obtain three types of reports: third molar, implant, and STL reports.
The third molar report is visually similar to the endodontic report. The doctor selects an area of interest – one of the wisdom teeth that is causing concern to the patient. The program itself selects the tooth, outlines the mandibular canal and shows vestibulo-oral, axial, mesio-distal sections with a slice thickness of 0.2 mm.
The implantology report independently calculates the width and length of the alveolar process. This necessary study of bone configuration is important for correct and predictable planning of the implantation result. Diagnocat provides an analysis of osseous conditions, taking into account the location of important anatomical structures such as the maxillary sinus and mandibular canal.
The STL report helps in creating templates for implantation. Diagnocat allows you to obtain fully automatic segmentation and create 3D STL models from DICOM files (CBCT) with simultaneous segmentation of individual structures (soft tissue, teeth, inferior alveolar canal, incisive canal, lower and upper jaw, skull base, respiratory tract).
With the rapid development of artificial intelligence, we are entering a new era of healthcare. Algorithms provide the opportunity to reduce routine work and focus our attention on critical selection of the optimal treatment strategy. However, it is important to remember that artificial intelligence still needs medical supervision. Perhaps in the near future there will be new functions for creating templates for implantation, modeling caps for indirect fixation, and perhaps someday it will be possible to simulate tooth movements in the same program. Developers do not stand still and are actively developing.
