RESUMO
Video-based heart and respiratory rate measurements using facial videos are more useful and user-friendly than traditional contact-based sensors. However, most of the current deep learning approaches require ground-truth pulse and respiratory waves for model training, which are expensive to collect. In this paper, we propose CalibrationPhys, a self-supervised video-based heart and respiratory rate measurement method that calibrates between multiple cameras. CalibrationPhys trains deep learning models without supervised labels by using facial videos captured simultaneously by multiple cameras. Contrastive learning is performed so that the pulse and respiratory waves predicted from the synchronized videos using multiple cameras are positive and those from different videos are negative. CalibrationPhys also improves the robustness of the models by means of a data augmentation technique and successfully leverages a pre-trained model for a particular camera. Experimental results utilizing two datasets demonstrate that CalibrationPhys outperforms state-of-the-art heart and respiratory rate measurement methods. Since we optimize camera-specific models using only videos from multiple cameras, our approach makes it easy to use arbitrary cameras for heart and respiratory rate measurements.
Assuntos
Taxa Respiratória , Autogestão , Humanos , Face , Coração , Frequência CardíacaRESUMO
BACKGROUND: Improved optical diagnostic technology is needed that can be used by also outside expert centers. Hence, we developed an artificial intelligence (AI) system that automatically and robustly predicts the pathological diagnosis based on the revised Vienna Classification using standard colonoscopy images. METHODS: We prepared deep learning algorithms and colonoscopy images containing pathologically proven lesions (56,872 images, 6775 lesions). Four classifications were adopted: revised Vienna Classification category 1, 3, and 4/5 and normal images. The best algorithm-ResNet152-in the independent internal validation (14,048 images, 1718 lesions) was used for external validation (255 images, 128 lesions) based on neoplastic and non-neoplastic classification. Diagnostic performance of endoscopists was compared using a computer-assisted interpreting test. RESULTS: In the internal validation, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy for adenoma (category 3) of 84.6% (95% CI 83.5-85.6%), 99.7% (99.5-99.8%), 90.8% (89.9-91.7%), 89.2% (88.5-99.0%), and 89.8% (89.3-90.4%), respectively. In the external validation, ResNet152's sensitivity, specificity, PPV, NPV, and accuracy for neoplastic lesions were 88.3% (82.6-94.1%), 90.3% (83.0-97.7%), 94.6% (90.5-98.8%), 80.0% (70.6-89.4%), and 89.0% (84.5-93.6%), respectively. This diagnostic performance was superior to that of expert endoscopists. Area under the receiver-operating characteristic curve was 0.903 (0.860-0.946). CONCLUSIONS: The developed AI system can help non-expert endoscopists make differential diagnoses of colorectal neoplasia on par with expert endoscopists during colonoscopy. (229/250 words).
Assuntos
Adenoma , Neoplasias Colorretais , Aprendizado Profundo , Humanos , Inteligência Artificial , Colonoscopia/métodos , Adenoma/diagnóstico , Adenoma/patologia , Neoplasias Colorretais/diagnóstico , Neoplasias Colorretais/patologiaRESUMO
Edema is a common symptom of kidney disease, and quantitative measurement of edema is desired. This paper presents a method to estimate the degree of edema from facial images taken before and after dialysis of renal failure patients. As tasks to estimate the degree of edema, we perform pre- and post-dialysis classification and body weight prediction. We develop a multi-patient pre-training framework for acquiring knowledge of edema and transfer the pre-trained model to a model for each patient. For effective pre-training, we propose a novel contrastive representation learning, called weight-aware supervised momentum contrast (WeightSupMoCo). WeightSupMoCo aims to make feature representations of facial images closer in similarity of patient weight when the pre- and post-dialysis labels are the same. Experimental results show that our pre-training approach improves the accuracy of pre- and post-dialysis classification by 15.1% and reduces the mean absolute error of weight prediction by 0.243 kg compared with training from scratch. The proposed method accurately estimate the degree of edema from facial images; our edema estimation system could thus be beneficial to dialysis patients.
RESUMO
Gaps in colonoscopy skills among endoscopists, primarily due to experience, have been identified, and solutions are critically needed. Hence, the development of a real-time robust detection system for colorectal neoplasms is considered to significantly reduce the risk of missed lesions during colonoscopy. Here, we develop an artificial intelligence (AI) system that automatically detects early signs of colorectal cancer during colonoscopy; the AI system shows the sensitivity and specificity are 97.3% (95% confidence interval [CI] = 95.9%-98.4%) and 99.0% (95% CI = 98.6%-99.2%), respectively, and the area under the curve is 0.975 (95% CI = 0.964-0.986) in the validation set. Moreover, the sensitivities are 98.0% (95% CI = 96.6%-98.8%) in the polypoid subgroup and 93.7% (95% CI = 87.6%-96.9%) in the non-polypoid subgroup; To accelerate the detection, tensor metrics in the trained model was decomposed, and the system can predict cancerous regions 21.9 ms/image on average. These findings suggest that the system is sufficient to support endoscopists in the high detection against non-polypoid lesions, which are frequently missed by optical colonoscopy. This AI system can alert endoscopists in real-time to avoid missing abnormalities such as non-polypoid polyps during colonoscopy, improving the early detection of this disease.
Assuntos
Colonoscopia , Neoplasias Colorretais/diagnóstico , Aprendizado Profundo , Processamento de Imagem Assistida por Computador/métodos , Colonoscopia/métodos , Sistemas Computacionais , Humanos , Estudos Retrospectivos , Sensibilidade e EspecificidadeRESUMO
We propose an algorithm for the detection of facial regions within input images. The characteristics of this algorithm are (1) a vast number of Gabor-type features (196,800) in various orientations, and with various frequencies and central positions, which are used as feature candidates in representing the patterns of an image, and (2) an information maximization principle, which is used to select several hundred features that are suitable for the detection of faces from among these candidates. Using only the selected features in face detection leads to reduced computational cost and is also expected to reduce generalization error. We applied the system, after training, to 42 input images with complex backgrounds (Test Set A from the Carnegie Mellon University face data set). The result was a high detection rate of 87.0%, with only six false detections. We compared the result with other published face detection algorithms.