A 3D and Explainable Artificial Intelligence Model for Evaluation of Chronic Otitis Media Based on Temporal Bone Computed Tomography: Model Development, Validation, and Clinical Application.

BACKGROUND: Temporal bone computed tomography (CT) helps diagnose chronic otitis media (COM). However, its interpretation requires training and expertise. Artificial intelligence (AI) can help clinicians evaluate COM through CT scans, but existing models lack transparency and may not fully leverage multidimensional diagnostic information. OBJECTIVE: We aimed to develop an explainable AI system based on 3D convolutional neural networks (CNNs) for automatic CT-based evaluation of COM. METHODS: Temporal bone CT scans were retrospectively obtained from patients operated for COM between December 2015 and July 2021 at 2 independent institutes. A region of interest encompassing the middle ear was automatically segmented, and 3D CNNs were subsequently trained to identify pathological ears and cholesteatoma. An ablation study was performed to refine model architecture. Benchmark tests were conducted against a baseline 2D model and 7 clinical experts. Model performance was measured through cross-validation and external validation. Heat maps, generated using Gradient-Weighted Class Activation Mapping, were used to highlight critical decision-making regions. Finally, the AI system was assessed with a prospective cohort to aid clinicians in preoperative COM assessment. RESULTS: Internal and external data sets contained 1661 and 108 patients (3153 and 211 eligible ears), respectively. The 3D model exhibited decent performance with mean areas under the receiver operating characteristic curves of 0.96 (SD 0.01) and 0.93 (SD 0.01), and mean accuracies of 0.878 (SD 0.017) and 0.843 (SD 0.015), respectively, for detecting pathological ears on the 2 data sets. Similar outcomes were observed for cholesteatoma identification (mean area under the receiver operating characteristic curve 0.85, SD 0.03 and 0.83, SD 0.05; mean accuracies 0.783, SD 0.04 and 0.813, SD 0.033, respectively). The proposed 3D model achieved a commendable balance between performance and network size relative to alternative models. It significantly outperformed the 2D approach in detecting COM (P≤.05) and exhibited a substantial gain in identifying cholesteatoma (P<.001). The model also demonstrated superior diagnostic capabilities over resident fellows and the attending otologist (P<.05), rivaling all senior clinicians in both tasks. The generated heat maps properly highlighted the middle ear and mastoid regions, aligning with human knowledge in interpreting temporal bone CT. The resulting AI system achieved an accuracy of 81.8% in generating preoperative diagnoses for 121 patients and contributed to clinical decision-making in 90.1% cases. CONCLUSIONS: We present a 3D CNN model trained to detect pathological changes and identify cholesteatoma via temporal bone CT scans. In both tasks, this model significantly outperforms the baseline 2D approach, achieving levels comparable with or surpassing those of human experts. The model also exhibits decent generalizability and enhanced comprehensibility. This AI system facilitates automatic COM assessment and shows promising viability in real-world clinical settings. These findings underscore AI's potential as a valuable aid for clinicians in COM evaluation. TRIAL REGISTRATION: Chinese Clinical Trial Registry ChiCTR2000036300; https://www.chictr.org.cn/showprojEN.html?proj=58685.

Impact of virtual reality training on mastoidectomy performance: a prospective randomised study.

PURPOSE: The opportunities for surgical training and practice in the operating room are in decline due to limited resources, increased efficiency demands, growing complexity of the cases, and concerns for patient safety. Virtual reality (VR) offers a novel opportunity to enhance surgical training and provide complementary three-dimensional experience that has been usually available in the operating room. Since VR allows viewing and manipulation of realistic 3D models, the VR environment could enhance anatomical and topographical knowledge, in particular. In this study, we explored whether incorporating VR anatomy training improves novices' performance during mastoidectomy over traditional methods. METHODS: Thirty medical students were randomized into two groups and taught mastoidectomy in a structured manner. One group utilized a VR temporal bone model during the training while the other group used more traditional materials such as anatomy books. After the training, all participants completed a mastoidectomy on a 3D-printed temporal bone model under expert supervision. Performance during the mastoidectomy was evaluated with multiple metrics and feedback regarding the two training methods was gathered from the participants. RESULTS: The VR training method was rated better by the participants, and they also needed less guidance during the mastoidectomy. There were no significant differences in operational time, the occurrence of injuries, self-assessment scores, and the surgical outcome between the two groups. CONCLUSION: Our results support the utilization of VR training in complete novices as it has higher trainee satisfaction and leads to at least as good results as the more traditional methods.

Risk factors for recurrent acute mastoiditis in pediatric patients: a registry-based cohort study.

OBJECTIVE: To describe characteristics of pediatric patients with recurrent acute mastoiditis, and to identify risk factors for this condition. STUDY DESIGN: A retrospective cohort study. SETTING: Data based on electronic medical records of the largest Health Maintenance Organization in Israel. METHODS: Children hospitalized due to acute mastoiditis during the years 2008-2018 were identified, and their diagnosis was verified. Patients with recurrent acute mastoiditis were identified and grouped, and their characteristics were outlined and compared to those of the original group to identify risk factors for recurrence. RESULTS: During the 11-year period, a total of 1115 cases of children hospitalized due to acute mastoiditis were identified with a weighted incidence rate of 7.8/100,000. Of this group, 57 patients were diagnosed with recurrence following a full clinical recovery. The incidence proportion of recurrent acute mastoiditis was 5.1% (57/1115), male-to-female ratio was 27:30, 73.4% were younger than 24 months, the median period from the first episode was 3.4 months (IQR 2.0;10.0), and 82.5% of the patients (n = 47) had a single recurrence, whereas 18.5% (n = 10) had two recurrences or more. Mastoidectomy and swelling over the mastoid area during the first episode were identified as the main risk factors for recurrent mastoiditis HR = 4.7 [(2.7-8.2), p < 0.001] and HR = 2.55 [(1.4-4.8), p = 0.003], respectively. Mastoidectomy was the only independent significant risk factor for recurrence in a multivariate analysis. CONCLUSIONS: Mastoidectomy and swelling over the mastoid area during the first episode of acute mastoiditis were found strongly related independent risk factor for future recurrent episodes of acute mastoiditis.

Applicability of ChOLE staging to preoperative HRCT temporal bone in chronic otitis media with cholesteatoma.

PURPOSE: Chronic otitis media with cholesteatoma is a frequent disease entity in otology, requiring surgery in overwhelming majority of cases. Despite the huge burden there is no established grading system available to assess the severity and extent of disease preoperatively until date. Aim of our study is to assess the applicability of ChOLE staging to preoperative HRCT temporal bone in Chronic otitis media with cholesteatoma. METHODOLOGY: Patients clinically diagnosed as COM with cholesteatoma, who underwent preoperative HRCT temporal bone imaging and mastoid exploration at our tertiary care centre were included. Preoperative radiology was assessed and a radiological ChOLE (r-ChOLE) was given by radiologist. This was then compared with the postop ChOLE. RESULTS: 21 patients were included in the study. Data was linear and normally distributed (Shapiro wilk test). Pearson's product-moment correlation used to see relationship between radiological and postop Total score showed strong statistically significant positive correlation with correlation coefficient (r) of 0.977. Paired t test showed p value was 0.329 (p > 0.05) suggesting no significant difference between radiological and postop Total scores. Cohen kappa test of agreement was applied. It revealed an overall strong agreement (p < 0.001). CONCLUSION: ChOLE staging may be extended to preoperative HRCT of temporal bone in COM with cholesteatoma (rCHOLE). A preoperative radiological staging will help in better prioritizing, planning and execution of tympanomastoid surgeries.

Rise in complications of acute otitis media during and after the COVID-19 pandemic.

PURPOSE: After the lifting of nonpharmaceutical interventions (NPIs) during the COVID-19 pandemic, clinical observation showed an increase in complications of acute otitis, followed by a rise in the number of mastoidectomies performed. The aim of this study was to record the number of mastoidectomies performed before, during and after the COVID-19 pandemic as an indicator for complications of acute otitis media. METHODS: Data were collected from a tertiary hospital in a university setting, as well as from four major public health insurance companies in Germany. The data of 24,824,763 German citizens during a period from 2014 until 2023 were analyzed. RESULTS: According to the data, during the COVID-19 pandemic, the number of mastoidectomies performed dropped by 54% for children aged 0-6 and by 62% for children aged 7-18. For adults, there were 30% fewer mastoidectomies performed between 2020 and 2022. After the lifting of most NPI's in the season from July 2022 to June 2023, there was a sharp increase in the number of mastoidectomies performed on patients of all ages. CONCLUSIONS: During the COVID-19 pandemic, a decrease in the number of mastoidectomies performed was seen, suggesting a lower incidence of complicated acute otitis, most likely linked to the general decrease of upper airway infections due to NPI's. In contrast, a sharp increase in the incidence of complicated otitis occurred after the hygiene measures were lifted. The current development causes a more frequent performance of mastoidectomies, thus entailing a change in the challenges for everyday clinical practice.

Surgical Option for External Auditory Canal Cholesteatoma: A Case Report.

External Auditory Canal Cholesteatomas (EACC), is an exceptionally rare condition with a prevalence of only 0.1-0.5% among new patients1. EACC are known to possess bone eroding properties, causing a variety of complications, similar to the better-known attic cholesteatomas. We describe here the novel surgical management of a case of EACC. She is 38-year-old female who presented with otorrhea for 6 months. Clinical examination and radiological investigations suggested the diagnosis of an external auditory canal cholesteatoma. The patient underwent modified radical mastoidectomy with type 1 tympanoplasty with meatoplasty. Post-operatively, the patient showed marked clinical improvement.

Characteristics of pediatric recurrent acute mastoiditis: A case-control study.

BACKGROUND: Data on pediatric recurrent acute mastoiditis are lacking, despite its morbidity and clinical significance. Our aim was to describe the incidence, characteristics, and associated factors of recurrent mastoiditis in hospitalized children. METHODS: Using a case-control design, analyzing electronic data of hospitalized children with acute mastoiditis between June 2011 and December 2018, children with recurrent mastoiditis were compared to children with a single episode at the time of hospitalization. Recurrent episodes of mastoiditis were compared to the first episodes. Recurrent acute mastoiditis was defined as recurring mastoiditis ≥4-weeks after a completely resolved event. RESULTS: Of 347 children hospitalized with acute mastoiditis, 22 (6.3%) had recurrent mastoiditis; the median interval between episodes was 3 months (range: 1-36). The mean ± SD age was 2.3 ± 2.25 years. A comparison of first episodes in recurring cases to single episodes by univariate and multivariate analysis, showed no differences in the pre-admission management or in the isolated pathogens; however, a history of atopic dermatitis and percutaneous abscess drainage were more frequent in first episodes of recurring cases (27.3% vs. 1.2%, p < 0.001, and 27.3% vs. 10.0%, p = 0.026, respectively). The second episode of acute mastoiditis was characterized by a milder clinical course and shorter durations from symptoms to hospitalization, intravenous antibiotic therapy, and length of hospital stay. Linear regression showed that an increased interval from symptoms to hospitalization significantly increased length of hospital stay (regression coefficient of 0.215 [95% CI: 0.114-0.317], p < 0.001). CONCLUSIONS: Recurrent episodes of mastoiditis were clinically milder, with shorter length of hospital stay compared to first episodes, possibly because of early admission.

Anterolateral (juxtacondylar) approach with limited mastoidectomy to resect a jugular foramen meningioma.

BACKGROUND: The anterolateral (juxtacondylar) approach with limited mastoidectomy is a suitable option to expose the postero-inferior part of the jugular foramen (JF). It is particularly indicated for tumors extending in the neck beyond the jugular foramen, especially in those cases necessitating both neck control as well as control of the mastoid segment of facial nerve. METHOD: We describe here the steps to safely perform an anterolateral approach with mastoidectomy along with a brief description of its indications and limits. CONCLUSION: This approach represents a valid option to reach the JF. Its knowledge can improve the process of optimal approach selection when dealing with complex pathology involving the JF.

Mastoid obliteration versus canal wall down or canal wall up mastoidectomy for cholesteatoma: Systematic review and meta-analysis.

BACKGROUND: There is ongoing debate over the optimal surgical approach for cholesteatoma treatment1-5. The aim of this review was to determine whether canal-wall down with mastoid obliteration is associated with favourable outcomes compared to either canal-wall down without obliteration or canal-wall up. The primary outcome was cholesteatoma reoccurrence with secondary outcomes including otorrhoea, reoperation and patient-reported outcome measures. METHODS: Medline, Embase, Cochrane databases and clinicaltrials.gov were searched for studies reporting outcomes of patients undergoing primary CWD-MO for cholesteatoma and directly compared with CWD and CWU mastoidectomies. Data were collated and meta-analysis performed. RESULTS: 2379 patients were included from 12 studies that met the inclusion criteria. CWD-MO was found to be associated with significantly less risk of recurrence when compared to CWU (OR = 0.330 (95 % CI 0.191-0.570, p < 0.001). When compared to CWD, CWD-MO was associate with significantly lower incidence of persistent otorrhoea (OR 0.405 (95 % CI 0.232-0.706, p < 0.001) while the rates of recurrence were comparable (OR: 1.231 [95 % CI: 0.550-2.757] p = 0.613). Otorrhoea rates between CWD-MO and CWU were equivalent as was reoperation. Quality of life and hearing outcomes were variably reported. CONCLUSIONS: This systematic review demonstrates that CWD-MO technique achieves lower recurrence rates comparable to CWD approaches and superior to CWU, while offering improved long-term control of otorrhoea. The final procedural choice should be decided based on surgeon experience, skill set and individual patient factors. To address the risk of selection bias, a randomised controlled trial is needed to answer this important research question.

Modified-modified radical mastoidectomy.

PURPOSE: It is unusual to have communication from the external auditory canal (EAC) directly to the mastoid, totally sparing the tympanum. These patients need a different surgical approach, a modified canal wall-down procedure, to completely clear the disease but fully preserve the tympanum. We present one such exceptional case. CASE PRESENTATION: A 28-year-old lady presented with ear discharge for 1 year. Imaging confirmed the canal-mastoid fistula, but the entire tympanum was normal. We performed a modified-modified radical mastoidectomy. CONCLUSIONS: Canal-mastoid fistula is an infrequent entity and may be idiopathic. Despite being evident on clinical examination, imaging aids in assessing size and location of the defect. Although EAC reconstruction may be attempted, the majority require a canal wall-down procedure.