Gland cancer is a high-incidence disease that endangers human health,and its early detection and treatment require efficient,accurate,and objective intelligent diagnosis methods.In recent years,the advent of machine l...Gland cancer is a high-incidence disease that endangers human health,and its early detection and treatment require efficient,accurate,and objective intelligent diagnosis methods.In recent years,the advent of machine learning techniques has yielded satisfactory results in intelligent gland cancer diagnosis based on clinical images,significantly improving the accuracy and efficiency of medical image interpretation while reducing the workload of doctors.The focus of this study is to review,classify,and analyze intelligent diagnosis methods for imaging gland cancer based on machine learning and deep learning.This paper briefly introduces some basic imaging principles of multimodal medical images,such as the commonly used computed tomography(CT),magnetic resonance imaging(MRI),ultrasound(US),positron emission tomography(PET),and pathology.In addition,the intelligent diagnosis methods for imaging gland cancer were further classified into supervised learning and weakly supervised learning.Supervised learning consists of traditional machine learning methods,such as K-nearest neighbor algorithm(KNN),support vector machine(SVM),and multilayer perceptron,and deep learning methods evolving from convolutional neural network(CNN).By contrast,weakly supervised learning can be further categorized into active learning,semisupervised learning,and transfer learning.State-of-the-art methods are illustrated with implementation details,including image segmentation,feature extraction,and optimization of classifiers.Their performances are evaluated through indicators,such as accuracy,precision,and sensitivity.In conclusion,the challenges and development trends of intelligent diagnosis methods for imaging gland cancer were addressed and discussed.展开更多
Objective: To review the records of cases of Bartholin’s Gland Carcinoma referred to the Queensland Centre for Gynaecological Cancer (QCGC) between mid 1993 and mid 2012. Methods: Bartholin’s Gland Carcinoma case da...Objective: To review the records of cases of Bartholin’s Gland Carcinoma referred to the Queensland Centre for Gynaecological Cancer (QCGC) between mid 1993 and mid 2012. Methods: Bartholin’s Gland Carcinoma case data from QCGC were reviewed and analysed using the computer software Statistical Package for the Social Sciences 11.0. Results: Of the 12 cases four died of their disease, seven are still alive and disease free and one is alive with recurrent disease. The mean age at diagnosis was 52.8 years. Time from onset of symptoms to diagnosis averaged 5.8 months. All diagnoses were confirmed histologically. Presenting symptoms included a lump and pain. The most common presenting complaint was a lump. Treatment included surgical excision, occasional biopsy followed by radiotherapy with or without chemotherapy. In some cases radiation and chemotherapy was followed by vulvectomy of various extent. Conclusions: Bartholin’s Gland Carcinoma is a rare condition with outcome dependent on duration of symptoms, including delay in diagnosis, cell-type, cellular differentiation and the International Federation of Gynecology and Obstetrics (FIGO) classification. A Bartholin’s gland mass in a woman aged 40 years or more should be considered malignant until a biopsy proves otherwise.The incidence of Bartholin’s Gland Carcinoma in Queensland is less than that reported elsewhere but a higher proportion of squamous cell carcinomas was found in this small series.展开更多
基金Supported by National Natural Science Foundation of China(62102036).
文摘Gland cancer is a high-incidence disease that endangers human health,and its early detection and treatment require efficient,accurate,and objective intelligent diagnosis methods.In recent years,the advent of machine learning techniques has yielded satisfactory results in intelligent gland cancer diagnosis based on clinical images,significantly improving the accuracy and efficiency of medical image interpretation while reducing the workload of doctors.The focus of this study is to review,classify,and analyze intelligent diagnosis methods for imaging gland cancer based on machine learning and deep learning.This paper briefly introduces some basic imaging principles of multimodal medical images,such as the commonly used computed tomography(CT),magnetic resonance imaging(MRI),ultrasound(US),positron emission tomography(PET),and pathology.In addition,the intelligent diagnosis methods for imaging gland cancer were further classified into supervised learning and weakly supervised learning.Supervised learning consists of traditional machine learning methods,such as K-nearest neighbor algorithm(KNN),support vector machine(SVM),and multilayer perceptron,and deep learning methods evolving from convolutional neural network(CNN).By contrast,weakly supervised learning can be further categorized into active learning,semisupervised learning,and transfer learning.State-of-the-art methods are illustrated with implementation details,including image segmentation,feature extraction,and optimization of classifiers.Their performances are evaluated through indicators,such as accuracy,precision,and sensitivity.In conclusion,the challenges and development trends of intelligent diagnosis methods for imaging gland cancer were addressed and discussed.
文摘Objective: To review the records of cases of Bartholin’s Gland Carcinoma referred to the Queensland Centre for Gynaecological Cancer (QCGC) between mid 1993 and mid 2012. Methods: Bartholin’s Gland Carcinoma case data from QCGC were reviewed and analysed using the computer software Statistical Package for the Social Sciences 11.0. Results: Of the 12 cases four died of their disease, seven are still alive and disease free and one is alive with recurrent disease. The mean age at diagnosis was 52.8 years. Time from onset of symptoms to diagnosis averaged 5.8 months. All diagnoses were confirmed histologically. Presenting symptoms included a lump and pain. The most common presenting complaint was a lump. Treatment included surgical excision, occasional biopsy followed by radiotherapy with or without chemotherapy. In some cases radiation and chemotherapy was followed by vulvectomy of various extent. Conclusions: Bartholin’s Gland Carcinoma is a rare condition with outcome dependent on duration of symptoms, including delay in diagnosis, cell-type, cellular differentiation and the International Federation of Gynecology and Obstetrics (FIGO) classification. A Bartholin’s gland mass in a woman aged 40 years or more should be considered malignant until a biopsy proves otherwise.The incidence of Bartholin’s Gland Carcinoma in Queensland is less than that reported elsewhere but a higher proportion of squamous cell carcinomas was found in this small series.
