Respiratory status classifying method and system thereof

A respiratory status classifying system, being for classifying as one of at least two respiratory statuses, and comprising:

• at least one processor;
• at least one program storage medium coupled to the processor and configured to provide a respiratory status classifier generating program;
• wherein the processor based on the respiratory status classifier generating program is configured to:
• process at least a part of a plurality of original physiological parameters, which are a plurality of physiological parameters for training, so as to generate a plurality of characteristic physiological parameters, wherein each of the original physiological parameters corresponds to a corresponding one of the respiratory statuses, the original physiological parameters comprise at least one clinical parameter, at least one ventilator parameter and at least one laboratory parameter, the clinical parameter comprises at least one of a body temperature, a respiration, a pulse, a systolic blood pressure, a blood oxygen and a urine output, the ventilator parameter comprises at least one of a fraction of inspired oxygen, a positive end-expiratory pressure, a peak airway pressure, a mean airway pressure, a respiratory rate and a tidal volume, the laboratory parameter comprises at least one of a procalcitonin, a partial pressure of arterial carbon dioxide and a partial pressure of arterial oxygen, each of the clinical parameter, the ventilator parameter and the laboratory parameter is an average value in a time period, which is between 12 hours and 72 hours, and the time period of each of the clinical parameter and the ventilator parameter is smaller than the time period of the laboratory parameter;
• process at least a part of a plurality of original chest images, which are a plurality of chest images for training, segment images of a left lung, a right lung and a heart from each of the original chest images that are processed, and enhance image data of the images being segmented, so as to generate a plurality of characteristic chest images, wherein each of the original chest images corresponds to a corresponding one of the respiratory statuses and is an X-ray chest image, and at least a part of the original chest images is a Chest X-ray 14 dataset published by National Institutes of Health, which has 15 types of chest symptoms labels being labeled thereon;
• train one of two respiratory status classifiers using the characteristic physiological parameters by a machine learning algorithm;
• train the other of the two respiratory status classifiers using the characteristic chest images by a CNN of DenseNet-121, wherein the characteristic chest images are labeled with the types of the chest symptoms labels;
• and
• generate the two respiratory status classifiers after training by the machine learning algorithm and the CNN, respectively, wherein the two respiratory status classifiers are a physiological parameter classifier and a chest image classifier, respectively, which are for classifying as one of the at least two respiratory statuses, and a last convolution block and a subsequent network weight of the CNN are adaptively updated for fine-tuning the chest image classifier according to relationships among the characteristic chest images, the at least two respiratory statuses and the 15 types of chest symptoms labels.