野田 昌生

BACKGROUND: The widespread adoption of smartphones and tablet devices, along with advancements in data communication technology, has resulted in a paradigm shift in the treatment of dizziness. External factors, such as the spread of COVID-19, have accelerated this transformation in recent years. We have been pursuing telemedicine and web-based medical care to treat dizziness and have developed different products and services necessary for each treatment process stage. Several patients face difficulties in accessing medical facilities during severe vertigo episodes. Furthermore, clinical findings, such as nystagmus or other symptoms, may be absent when symptoms subside by the time of their appointment. OBJECTIVE: This study aimed to develop a smartphone app for capturing eye movements and head positions during vertigo attacks, enabling recordings anywhere, even at home or work. METHODS: We developed an app named "iCapNYS" that uses the iPhone's front camera and gyro sensor to record eye movements and head positions. The app incorporates features designed to encourage spontaneous eye movements, minimizing nystagmus suppression caused by fixation. Additionally, we designed lightweight, recyclable cardboard goggles to securely hold the smartphone and block visual stimuli from the surrounding environment, optimizing the recording conditions. RESULTS: The "iCapNYS" system successfully captured subtle peripheral vestibular nystagmus in a patient with vertigo. The recorded nystagmus characteristics are comparable to those obtained using traditional infrared CCD (charge-coupled device) cameras. CONCLUSIONS: This app is an effective tool for treating vertigo and is easy for older adults to use, as it can be recorded with only 3 taps. We expect that the introduction of this nystagmus-monitoring system will improve vertigo treatment quality, promote medical collaboration, and provide patients with peace of mind in their care.

BACKGROUND: Conventional nystagmus classification methods often rely on subjective observation by specialists, which is time-consuming and variable among clinicians. Recently, deep learning techniques have been used to automate nystagmus classification using convolutional and recurrent neural networks. These networks can accurately classify nystagmus patterns using video data. However, associated challenges including the need for large datasets when creating models, limited applicability to address specific image conditions, and the complexity associated with using these models. OBJECTIVE: This study aimed to evaluate a novel approach for nystagmus classification that used the Generative Pre-trained Transformer 4 Vision (GPT-4V) model, which is a state-of-the-art large-scale language model with powerful image recognition capabilities. METHODS: We developed a pupil-tracking process using a nystagmus-recording video and verified the optimization model's accuracy using GPT-4V classification and nystagmus recording. We tested whether the created optimization model could be evaluated in six categories of nystagmus: right horizontal, left horizontal, upward, downward, right torsional, and left torsional. The traced trajectory was input as two-dimensional coordinate data or an image, and multiple in-context learning methods were evaluated. RESULTS: The developed model showed an overall classification accuracy of 37% when using pupil-traced images and a maximum accuracy of 24.6% when pupil coordinates were used as input. Regarding orientation, we achieved a maximum accuracy of 69% for the classification of horizontal nystagmus patterns but a lower accuracy for the vertical and torsional components. CONCLUSIONS: We demonstrated the potential of versatile vertigo management in a generative artificial intelligence model that improves the accuracy and efficiency of nystagmus classification. We also highlighted areas for further improvement, such as expanding the dataset size and enhancing input modalities, to improve classification performance across all nystagmus types. The GPT-4V model validated only for recognizing still images can be linked to video classification and proposed as a novel method.

BACKGROUND: Paediatric cervical abscesses necessitate careful assessment to determine appropriate treatment strategies. Some patients require surgical intervention, although conservative management is effective. However, the criteria for the surgical indications remain unclear. Machine learning models have demonstrated promise in improving diagnostic accuracy across different medical fields. OBJECTIVE: This study aimed to assess the use of machine learning models in predicting the requirement for surgical intervention in paediatric cervical abscesses and compare their performance with that of traditional logistic regression. METHODS: A retrospective analysis was conducted on 55 paediatric patients diagnosed with cervical abscesses between 2010 and 2024. The patient demographics, clinical findings, laboratory data, and imaging characteristics were examined. Six predictive models were developed: logistic regression, Random Forest, Lasso regression, Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine. Model performance was evaluated using the area under the curve (AUC), accuracy, precision, recall, and F1-score. Feature importance was examined to identify the main predictive factors. RESULTS: Among all the factors, abscess size was the most significant predictor of surgical intervention. Machine-learning models, especially XGBoost, outperformed logistic regression, achieving the highest AUC, accuracy, and recall. Inflammatory markers, including neutrophil-to-lymphocyte ratio and neutrophil count, also substantially contributed to the prediction accuracy. CONCLUSION: Machine learning models, particularly XGBoost, provide superior predictive performance compared with logistic regression, providing a valuable tool for optimising treatment decisions in paediatric cervical abscesses. These models improve clinical decision-making by integrating multiple factors, decreasing unnecessary surgeries, and enhancing patient outcomes.

BACKGROUND: Observing eye movements during episodic vertigo attacks is crucial for accurately diagnosing vestibular disorders. In clinical practice, many cases lack observable symptoms or clear findings during outpatient examinations, leading to diagnostic challenges. An accurate diagnosis is essential for timely treatment, as conditions such as benign paroxysmal positional vertigo (BPPV), Ménière's disease, and vestibular migraine require different therapeutic approaches. OBJECTIVE: This study aimed to develop and evaluate a cost-effective diagnostic tool that integrates a mini-infrared camera with 3D-printed goggles, enabling at-home recording of nystagmus during vertigo attacks. METHODS: A commercially available mini-infrared camera (US $25) was combined with 3D-printed goggles (US $13) to create a system for recording eye movements in dark conditions. A case study was conducted on a male patient in his 40s who experienced recurrent episodic vertigo. RESULTS: Initial outpatient evaluations, including oculomotor and vestibular tests using infrared Frenzel glasses, revealed no spontaneous or positional nystagmus. However, with the proposed system, the patient successfully recorded geotropic direction-changing positional nystagmus during a vertigo attack at home. The nystagmus was beating distinctly stronger on the left side down with 2.0 beats/second than the right side down with 1.2 beats/second. Based on the recorded videos, a diagnosis of lateral semicircular canal-type BPPV was made. Treatment with the Gufoni maneuver effectively alleviated the patient's symptoms, confirming the diagnosis. The affordability and practicality of the device make it particularly suitable for telemedicine and emergency care applications, enabling patients in remote or underserved areas to receive accurate diagnoses. CONCLUSIONS: The proposed system demonstrates the feasibility and utility of using affordable, accessible technology for diagnosing vestibular disorders outside of clinical settings. By addressing key challenges, such as the absence of symptoms during clinical visits and the high costs associated with traditional diagnostic tools, this device offers a practical solution for real-time monitoring and accurate diagnosis. Its potential applications extend to telemedicine, emergency settings, and resource-limited environments. Future iterations that incorporate higher-resolution imaging and automated analysis could further enhance its diagnostic capabilities and usability across diverse patient populations.

BACKGROUND: Reduction of spontaneous nystagmus by fixation, a characteristic feature of peripheral nystagmus, is important for differentiating between peripheral and central vestibular disorders. In the emergency room, Frenzel goggles are recommended to observe spontaneous nystagmus for the differential diagnosis of acute vestibular syndrome. We developed a portable loupe with a Fresnel lens to observe nystagmus. The loupe does not require power supply and can be used under ceiling lights. The aim of this study was to quantitatively and objectively compare the abilities of the loupe and conventional Frenzel goggles to observe spontaneous nystagmus and to verify that the loupe can detect peripheral nystagmus that cannot be observed with the naked eye. METHODS: Visual impact susceptibility was compared between the loupe and Frenzel goggles using the slow-phase velocity of nystagmus induced by the caloric test in 15 participants. Subsequently, under lighting, the nystagmus observations under the naked eye condition and with the use of the loupe were compared. Furthermore, the visibility of nystagmus was evaluated from recorded videographic images. RESULTS: In observations of nystagmus induced by the caloric test, the visual impact of the loupe was not inferior to that of Frenzel goggles. The mean slow-phase velocity of nystagmus recorded with the loupe was significantly higher than that observed with the naked eye. Nystagmus weakened under bright lighting could be recovered by the loupe as fixation was blocked and the direction of the nystagmus could be defined. CONCLUSIONS: The results showed that the loupe is helpful in observing nystagmus, which is weakly observed with the naked eye under bright light. This portable, low-cost loupe, which yields superior results, can serve as an alternative to conventional Frenzel goggles in emergency medical settings where rapid assessment is required.