• 사상체질의학회지
• /
• 제18권3호
• /
• pp.75-93
• /
• 2006

1. Objectives and Methods Sasang Constitutional Medicine is the original Korean Medicine. The purpose of this study was to objectify the diagnosis of Sasang Constitution. 212 Women's sentences were analyzed into 228 factors like Pitch, APQ, Shimmer, Octave and Energy, etc. Women's sentences were classified into 3 categories: total group, under 54 years old group and over 55 years old group. 2. Results 1) In Total group Soyangin's Center feq.(3) was significantly high compared with Taeyangin and Taeumin groups. Taeumin's Pitch2 was significantly high compared with Soeumin and Taeyangin groups. Taeyangin's Pitch S.D. was significantly high compared with Soyangin group. Taeyangin's Octave6 was significantly high compared with Soeumin group. There were no significant differences among constitutional groups in APQ and Shimmer segment. On the point of Energy, Taeyangin's G Tot E(1), G# Tot E(1), G dev.(1), G# dev.(1), G Tot E(2), G# Tot E(2), G dev.(4) and G# dev.(4) were significantly high compared with other groups. Soyangin's A#S.D.(2) was significantly high compared with Taeyangin group. Taeyangin's A#S.D.(3) was significantly high compared with Taeumin group. Taeyangin's F S.D.(5), F# S.D.(5) and Max Average were significantly high compared with Soeumin group. Taeumin's Peak3 and Peak4 were significantly high compared with Taeyangin group. Taeumin's PeakValue1 was significantly high compared with Soeumin group. Taeyangin's PeakValue2 was significantly high compared with Soeumin group. Taeyangin's PeakValue3 and PeakValue5 were significantly high compared with Other groups. 2) In Under 54 years old group, there were no significant differences among constitutional groups in APQ, Shimmer and Octave segment. Taeumin's Center freq.(2) was significantly high compared with Taeyangin and Soyangin groups. Taeumin's Pitch(2) and Pitch(3) were significantly high compared with Taeyangin and Soeumin groups. Taeyangin's and Taeumin's Pitch S.D. were significantly high compared with Soyangin group. Taeyangin's and Soyangin's Octave2 were significantly high compared with Taeumin group. On the point of Energy, Taeyangin's and Soyangin's A# S.D.(2) were significantly high compared with Soeumin group. Taeyangin's and Soyangin's G# dev.(1), G# dev.(2) were significantly high compared with Taeumin group. Taeyangin's and Taeumin's F# S.D.(3) were significantly high compared with Soeumin group. Taeyangin's and Soyangin's Max Average were significantly high compared with Soeumin group. Taeumin's Peak3 was significantly high compared with Taeyangin and Soeumin groups. Taeyangin's and Taeumin's PeakValue2 were significantly high compared with Soeumin group. Taeyangin's and Soeumin's PeakValue3 were significantly high compared with Taeumin group. Taeyangin's and Soyangin's PeakValue5 were significantly high compared with Soeumin group. Taeyangin's and Soyangin's PeakValue9 were significantly high compared with Taeumin group 3) In Over 55 years old group, there were no significant differences among constitutional groups in Pitch, APQ, and Peak segment. Soeumin's F Shimmer(1) and F Shimmer(2) were significantly high compared with Taeyangin and Taeumin groups. Soeumin's G# Shimmer(1) and G# Shimmer(2) were significantly high compared with Soyangin group. Taeyangin's Octave5 and Octave6 were significantly high compared with Soeumin group. On the point of Energy, Soyangin's C S.D., F# S.D.(1), F# S.D.(2) and G dev.(2) were significantly high compared with other groups. Soyangin's F# S.D.(3) was significantly high compared with Taeumin and Soeumin groups. Taeyangin's and Taeumin's G# S.D.(2) and G# S.D.(3) were significantly high compared with Soyangin group 3. Conclusions From above result, there is the possibility of efficient standard guide for constitution diagnosis by analysis of voice

• 지능정보연구
• /
• 제29권3호
• /
• pp.57-78
• /
• 2023

2019년 11월 중국 우한시에서 발병한 코로나19는 2020년 중국을 넘어 세계로 퍼져나가 2020년 3월에는 전 세계적으로 확산되었다. 코로나19와 같이 전염성이 강한 바이러스는 예방과 확진시 적극적인 치료도 중요하지만 우선 전파 속도가 빠른 바이러스인 점을 감안할 때, 확진 사실을 재빠르게 파악하여 전파를 차단하는 것이 더욱 중요하다. 그러나 감염여부를 확인하기 위한 PCR검사는 비용과 시간이 많이 소요되고, 자가키트검사 또한 접근성은 쉽지만 매번 수시로 받기에는 키트의 가격이 부담이 될 수밖에 없는 실정이다. 이러한 상황에서 기침 소리를 기반으로 코로나19 양성 여부를 판단할 수 있게 된다면 누구나 쉽게 언제, 어디서든 확진 여부를 체크할 수 있어 신속성과 경제성 측면에서 큰 장점을 가질 수 있을 것이다. 따라서 본 연구는 기침 소리를 기반으로 코로나19 확진 여부를 식별할 수 있는 분류 모델을 개발하는 것을 목적으로 하였다. 이를 위해, 본 연구에서는 먼저 MFCC, Mel-Spectrogram, Spectral contrast, Spectrogram 등을 통해 기침 소리를 벡터화 하였다. 이 때, 기침 소리의 품질을 위해 SNR을 통해 잡음이 많은 데이터는 삭제하였고, chunk를 통해 음성 파일에서 기침 소리만 추출하였다. 이후, 추출된 기침 소리의 feature를 이용하여 코로나 양성과 음성을 분류하기 위한 모델을 구축하였으며, XGBoost, LightGBM, FCNN 알고리즘을 통해 모델 학습을 수행하고 각 알고리즘별 성능을 비교하였다. 또한, 기침 소리를 다차원 벡터로 변환한 경우와, 이미지로 변환한 경우에 대해 모델 성능에 대한 비교 실험을 수행하였다. 실험 결과, 건강상태에 대한 기본정보와 기침 소리를 MFCC, Mel-Spectogram, Spectral contrast, 그리고 Spectrogram을 통해 다차원 벡터로 변환한 feature를 모두 활용한 LightGBM 모델이 0.74의 가장 높은 정확도를 보였다.