Typically, classification is conducted on a dataset that consists of numerical features and target classes. For instance, a grayscale image, which is usually represented as a matrix of integers varying from 0 to 255, enables one to apply various classification algorithms to image classification tasks. However, datasets represented as binary features cannot use many standard machine learning algorithms optimally, yet their amount is not negligible. On the other hand, oversampling algorithms such as synthetic minority oversampling technique (SMOTE) and its variants are often used if the dataset for classification is imbalanced. However, since SMOTE and its variants synthesize new minority samples based on the original samples, the diversity of the samples synthesized from binary features is highly limited due to the poor representation of original features. To solve this problem, a preprocessing approach is studied. By converting binary features into numerical ones using feature extraction methods, succeeding oversampling methods can fully display their potential in improving the classifiers’ performances. Through comprehensive experiments using benchmark datasets and real medical datasets, it was observed that a converted dataset consisting of numerical features is better for oversampling methods (maximum improvements of accuracy and F1-score were 35.11% and 42.17%, respectively). In addition, it is confirmed that feature extraction and oversampling synergistically contribute to the improvement of classification performance.

Globally, syphilis cases are rising, and varied symptoms hinder diagnosis, highlighting the role of serological testing. Comprehensive institutional analysis is needed, as many reports are limited. We retrospectively reviewed 23 persons with syphilis treated at Kanazawa University Hospital (January 2007–December 2023). Of 9145 individuals tested for treponemal (chemiluminescent enzyme immunoassay) and nontreponemal (rapid plasma reagin, RPR) antibodies, diagnoses were based on clinical assessment and serology. Data on history, stage, lesions, and treatment were collected. From 2007 to 2023, antibody testing increased, but treated cases stayed stable (0–4/year). Males comprised 60.9% and females 39.1%. In the treated 23 patients, ophthalmology (30.4%) and dermatology (17.4%) accounted for nearly half of test requests, though cases arose across specialties. Early-stage and ocular syphilis each occurred in 34.8%. Uveitis was the most common lesion (26.1%). Treatment included amoxicillin (69.6%) or penicillin (21.7%). At diagnosis, 78.3% were dual-positive and 21.7% treponemal-only positive. Nine patients (64.3%) had a ≥4-fold RPR titer decline (median 143.5 days); no RPR increases suggested treatment failure or reinfection. This study found stable treated numbers, with ocular syphilis and uveitis frequent. In specialized institutions, clinicians should consider syphilis in persons with varied symptoms across departments, ensuring comprehensive testing and appropriate follow-up.