Rainer Schmelzeisen’s research while affiliated with University of Freiburg and other places

BACKGROUND Podcasts are gaining popularity in health professions education due to their accessibility, flexibility, and potential to foster clinical reasoning. However, most formats remain asynchronous and non-interactive, and evidence-based, didactically grounded implementations in dental education are lacking. OBJECTIVE This study aimed to design, implement, and evaluate a novel teaching format - the Live Podcast (LP) - to enhance interdisciplinary thinking and clinical reasoning in dentomaxillofacial radiology. Both objective learning outcomes and student perceptions were assessed. METHODS A total of 41 undergraduate dental students from two cohorts participated. The intervention group (n = 21) engaged in a LP session featuring an interdisciplinary case discussion between a dental radiologist and a clinical specialist. The control group (n = 20) received no podcast intervention. Knowledge acquisition was measured via post-session test. Additionally 21 students completed a structured evaluation questionnaire. Descriptive and inferential statistics (Chi², Fisher’s exact test, Mann–Whitney U test) were used for data analysis. RESULTS Students in the intervention group scored significantly higher in the domain of operative dentistry (mean 4.62 vs. 3.75, p = 0.015), while overall performance across all topics showed a non-significant advantage of the intervention group. The LP format received high ratings (mean value out of 10) across all evaluation domains, including structure (9.76), interactivity (9.62), and interdisciplinary relevance (9.55). Students expressed a strong interest in continued use and recommended the format as a valuable addition to traditional teaching. CONCLUSIONS LP offers an engaging and learner-centered approach to dental education. By integrating expert dialogue, real-time interaction, and case-based learning, it fosters clinical reasoning and promotes interdisciplinary awareness. The high level of learner acceptance highlights its potential as an effective and meaningful educational tool. Further research is warranted to assess its scalability, sustainability, and long-term impact across institutions and dental specialties.

Objectives This study evaluated the accuracy and reliability of deep learning-based segmentation techniques for mandibular canal identification in CBCT data to provide a reliable and efficient support-tool for dental implant treatment planning. Methods A dataset of 90 cone beam computed tomography (CBCT) scans was annotated as ground truth for mandibular canal segmentation. The dataset was split into training (n = 69), validation (n = 1), and testing (n = 20) subsets. A deep learning model based on a hierarchical convolutional neural network architecture was developed and trained. The model's performance was evaluated using Dice similarity coefficient (DSC), 95% Hausdorff distance (HD), and average symmetric surface distance (ASSD). Qualitative assessment was performed by two experienced dental imaging practitioners who evaluated the segmentation quality in terms of trust and safety on a 5-point Likert scale at three mandibular locations per side. Results The trained model achieved a mean DSC of 0.77 ± 0.09, HD of 1.66 ± 0.86 mm, and ASSD of 0.31 ± 0.15 mm on the testing subset. Qualitative assessment showed no significant difference between the deep learning-based segmentations and ground truth in terms of trust and safety across all investigated locations (p > 0.05). Conclusions The proposed deep learning-based segmentation technique exhibits sufficient accuracy for the reliable identification of mandibular canals in CBCT scans. This automated approach could streamline the pre-operative planning process for dental implant placement, reducing the risk of neurovascular complications and enhancing patient safety.

This experimental phantom study investigates current standard of care protocols in cone beam computed tomography (CBCT), energy-integrating-detector (EID) CT, and photon-counting-detector (PCD) CT regarding their potential in delineation of dental root canals. Artificial accessory canals (diameters: 1000, 600, 400, 300 and 200 μm) were drilled into three bovine teeth mounted on a bovine rib as a jaw substitute. The phantom was scanned in two dental CBCTs, two EID-CTs and a PCD-CT using standard clinical protocols. Scans from a micro-CT served as reference standard. Spatial resolution was evaluated via line profiles through the canals, whereby visibility compared to surrounding noise and compared to the ground truth were assessed. PCD-CT was able to delineate all artificial canals down to 200 μm diameter. In CBCT and EID-CT canals could only be reliably detected down to 300 μm. Also, PCD-CT showed a considerably smaller width-divergence from the ground trough with 4.4% at 1000 μm and 35.1% at 300 μm compared to CBCT (13.5 and 72.9%) and EID-CT (10.1 and 115.7%). PCD-CT provided superior resolution, accurate size measurement, and enhanced detection of small dental root canals, thereby offering improvements in diagnostic capabilities compared to CBCT and EID-CT systems.

Computer technology–based treatment approaches like intraoperative navigation and intensity-modulated radiation therapy have become important components of state of the art head and neck cancer treatment. Multidirectional exchange of virtual three-dimensional patient data via an interdisciplinary platform allows all medical specialists involved in the patients treatment to take full advantage of these technologies. This review article gives an overview of current technologies and future directions regarding treatment approaches that are based on a virtual, three-dimensional patient specific dataset: storage and exchange of spatial information acquired via intraoperative navigation allow for a highly precise frozen section procedure. In the postoperative setting, virtual reconstruction of the tumor resection surface provides the basis for improved radiation therapy planning and virtual reconstruction of the tumor with integration of molecular findings creates a valuable tool for postoperative treatment and follow-up. These refinements of established treatment components and novel approaches have the potential to make a major contribution to improving the outcome in head and neck cancer patients.

Stimulated Raman Histology (SRH) employs the stimulated Raman scattering (SRS) of photons at biomolecules in tissue samples to generate histological images. Subsequent pathological analysis allows for an intraoperative evaluation without the need for sectioning and staining. The objective of this study was to investigate a deep learning-based classification of oral squamous cell carcinoma (OSCC) and the sub-classification of non-malignant tissue types, as well as to compare the performances of the classifier between SRS and SRH images. Raman shifts were measured at wavenumbers k1 = 2845 cm−1 and k2 = 2930 cm−1. SRS images were transformed into SRH images resembling traditional H&E-stained frozen sections. The annotation of 6 tissue types was performed on images obtained from 80 tissue samples from eight OSCC patients. A VGG19-based convolutional neural network was then trained on 64 SRS images (and corresponding SRH images) and tested on 16. A balanced accuracy of 0.90 (0.87 for SRH images) and F1-scores of 0.91 (0.91 for SRH) for stroma, 0.98 (0.96 for SRH) for adipose tissue, 0.90 (0.87 for SRH) for squamous epithelium, 0.92 (0.76 for SRH) for muscle, 0.87 (0.90 for SRH) for glandular tissue, and 0.88 (0.87 for SRH) for tumor were achieved. The results of this study demonstrate the suitability of deep learning for the intraoperative identification of tissue types directly on SRS and SRH images.

Objectives To develop a content-aware chatbot based on GPT-3.5-Turbo and GPT-4 with specialized knowledge on the German S2 Cone-Beam CT (CBCT) dental imaging guideline and to compare the performance against humans. Methods The LlamaIndex software library was used to integrate the guideline context into the chatbots. Based on the CBCT S2 guideline, 40 questions were posed to content-aware chatbots and early career and senior practitioners with different levels of experience served as reference. The chatbots’ performance was compared in terms of recommendation accuracy and explanation quality. Chi-square test and one-tailed Wilcoxon signed rank test evaluated accuracy and explanation quality, respectively. Results The GPT-4 based chatbot provided 100% correct recommendations and superior explanation quality compared to the one based on GPT3.5-Turbo (87.5% vs. 57.5% for GPT-3.5-Turbo; p = 0.003). Moreover, it outperformed early career practitioners in correct answers (p = 0.002 and p = 0.032) and earned higher trust than the chatbot using GPT-3.5-Turbo (p = 0.006). Conclusions A content-aware chatbot using GPT-4 reliably provided recommendations according to current consensus guidelines. The responses were deemed trustworthy and transparent and therefore facilitate the integration of artificial intelligence into clinical decision-making.

Purpose: The aim of the study was to assess the deviation between clinical implant axes (CIA) determined by a surgeon during preoperative planning and reconstructed tooth axes (RTA) of missing teeth which were automatically computed by a previously introduced anatomical SSM. Methods: For this purpose all available planning datasets of single-implant cases of our clinic, which were planned with coDiagnostix Version 9.9 between 2018 and 2021, were collected for retrospective investigation. Informed consent was obtained. First, the intraoral scans of implant patients were annotated and subsequently analyzed using the SSM. The RTA, computed by the SSM, was then projected into the preoperative planning dataset. The amount and direction of spatial deviation between RTA and CIA were then measured. Results: Thirty-five patients were implemented. The mean distance between the occlusal entry point of anterior and posterior implants and the RTA was 0.99 mm ± 0.78 mm and 1.19 mm ± 0.55, respectively. The mean angular deviation between the CIA of anterior and posterior implants and the RTA was 12.4° ± 3.85° and 5.27° ± 2.97° respectively. The deviations in anterior implant cases were systematic and could be corrected by computing a modified RTA (mRTA) with decreased deviations (0.99 mm ± 0.84 and 4.62° ± 1.95°). The safety distances of implants set along the (m)RTA to neighboring teeth were maintained in 30 of 35 cases. Conclusion: The RTA estimated by the SSM revealed to be a viable implant axis for most of the posterior implant cases. As there are natural differences between the anatomical tooth axis and a desirable implant axis, modifications were necessary to correct the deviations which occurred in anterior implant cases. However, the presented approach is not applicable for clinical use and always requires manual optimization by the planning surgeon.

Objectives: This study evaluated the accuracy of deep neural patchworks (DNPs), a deep learning-based segmentation framework, for automated identification of 60 cephalometric landmarks (bone-, soft tissue- and tooth-landmarks) on CT scans. The aim was to determine whether DNP could be used for routine three-dimensional cephalometric analysis in diagnostics and treatment planning in orthognathic surgery and orthodontics. Methods: Full skull CT scans of 30 adult patients (18 female, 12 male, mean age 35.6 years) were randomly divided into a training and test data set (each n = 15). Clinician A annotated 60 landmarks in all 30 CT scans. Clinician B annotated 60 landmarks in the test data set only. The DNP was trained using spherical segmentations of the adjacent tissue for each landmark. Automated landmark predictions in the separate test data set were created by calculating the center of mass of the predictions. The accuracy of the method was evaluated by comparing these annotations to the manual annotations. Results: The DNP was successfully trained to identify all 60 landmarks. The mean error of our method was 1.94 mm (SD 1.45 mm) compared to a mean error of 1.32 mm (SD 1.08 mm) for manual annotations. The minimum error was found for landmarks ANS 1.11 mm, SN 1.2 mm, and CP_R 1.25 mm. Conclusion: The DNP-algorithm was able to accurately identify cephalometric landmarks with mean errors <2 mm. This method could improve the workflow of cephalometric analysis in orthodontics and orthognathic surgery. Low training requirements while still accomplishing high precision make this method particularly promising for clinical use.

Since its discovery in 1989, RNA interference (RNAi) has become a widely used tool for the in vitro downregulation of specific gene expression in molecular biological research. This basically involves a complementary RNA that binds a target sequence to affect its transcription or translation process. Currently, various small RNAs, such as small interfering RNA (siRNA), micro RNA (miRNA), small hairpin RNA (shRNA), and PIWI interacting RNA (piRNA), are available for application on in vitro cell culture, to regulate the cells’ gene expression by mimicking the endogenous RNAi-machinery. In addition, several biochemical, physical, and viral methods have been established to deliver these RNAs into the cell or nucleus. Since each RNA and each delivery method entail different off-target effects, limitations, and compatibilities, it is crucial to understand their basic mode of action. This review is intended to provide an overview of different nucleic acids and delivery methods for planning, interpreting, and troubleshooting of RNAi experiments.

Objectives To investigate whether in patients undergoing surgery for oral squamous cell carcinoma, stimulated Raman histology (SRH), in comparison with H&E-stained frozen sections, can provide accurate diagnoses regarding neoplastic tissue and sub-classification of non-neoplastic tissues. Materials and methods SRH, a technology based on Raman scattering, was applied to generate digital histopathologic images of 80 tissue samples obtained from 8 oral squamous cell carcinoma (OSCC) patients. Conventional H&E-stained frozen sections were then obtained from all 80 samples. All images/sections (SRH and H&E) were analyzed for squamous cell carcinoma, normal mucosa, connective tissue, muscle tissue, adipose tissue, salivary gland tissue, lymphatic tissue, and inflammatory cells. Agreement between SRH and H&E was evaluated by calculating Cohen’s kappa. Accuracy of SRH compared to H&E was quantified by calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) as well as area under the receiver operating characteristic curve (AUC). Results Thirty-six of 80 samples were classified as OSCC by H&E-based diagnosis. Regarding the differentiation between neoplastic and non-neoplastic tissue, high agreement between H&E and SRH (kappa: 0.880) and high accuracy of SRH (sensitivity: 100%; specificity: 90.91%; PPV: 90.00%, NPV: 100%; AUC: 0.954) were demonstrated. For sub-classification of non-neoplastic tissues, SRH performance was dependent on the type of tissue, with high agreement and accuracy for normal mucosa, muscle tissue, and salivary glands. Conclusion SRH provides high accuracy in discriminating neoplastic and non-neoplastic tissues. Regarding sub-classification of non-neoplastic tissues in OSCC patients, accuracy varies depending on the type of tissue examined. Clinical relevance This study demonstrates the potential of SRH for intraoperative imaging of fresh, unprocessed tissue specimens from OSCC patients without the need for sectioning or staining.