• The Journal of the Acoustical Society of Korea
• /
• v.20 no.4
• /
• pp.35-41
• /
• 2001

This paper describes Linear Discriminant Analysis and common vector extraction for speech recognition. Voice signal contains psychological and physiological properties of the speaker as well as dialect differences, acoustical environment effects, and phase differences. For these reasons, the same word spelled out by different speakers can be very different heard. This property of speech signal make it very difficult to extract common properties in the same speech class (word or phoneme). Linear algebra method like BT (Karhunen-Loeve Transformation) is generally used for common properties extraction In the speech signals, but common vector extraction which is suggested by M. Bilginer et at. is used in this paper. The method of M. Bilginer et al. extracts the optimized common vector from the speech signals used for training. And it has 100％ recognition accuracy in the trained data which is used for common vector extraction. In spite of these characteristics, the method has some drawback-we cannot use numbers of speech signal for training and the discriminant information among common vectors is not defined. This paper suggests advanced method which can reduce error rate by maximizing the discriminant information among common vectors. And novel method to normalize the size of common vector also added. The result shows improved performance of algorithm and better recognition accuracy of 2％ than conventional method.

• Journal of the Korean Arthroscopy Society
• /
• v.17 no.1
• /
• pp.31-37
• /
• 2013

Purpose: This study performed to compare degree of joint stiffness and clinical results between early and delayed reconstruction of acute posterior cruciate ligament (PCL) injuries. Materials and Methods: Thirty-two Patients who underwent PCL reconstruction between March 2008 and October 2011 enrolled this study. We performed transtibial single bundle reconstruction using the allo-achilles tendon in all cases. We divided the patient into two groups, early reconstruction group underwent surgery before a week, delayed reconstruction group underwent surgery after 3 weeks, before 6 weeks. All the patients underwent aggressive joint motion exercise till surgery and enrolled post operative rehabilitation program. We checked posterior drawer stress radiography, range of motion, the Lysholm score, the International Knee Documentation Committee (IKDC) score and the Tegner score to evaluate the results. Results: At the final follow up the Lysholm score was 92.1 in the early group and 93.8 in the delayed group. All the cases were rated above B (near normal) on IKDC score (p=0.808, p=0.722). The Tegner score was 6.6 in the early reconstruction group and 6.2 in the delayed group (p=0.480), The average of maximum flexion and extension angle was $133.9^{\circ}$, $1.4^{\circ}$ in the early group and $133.6^{\circ}$, $1.1^{\circ}$ in the delayed group (p=0.560, p=0.581), no complication such as deep vein thrombosis or infection, no difference in posterior drawer stress radiography (p=0.750). Conclusion: We could obtain satisfactory clinical results in both the early and delayed reconstruction groups of acute PCL injuries. Therefore, the early reconstruction of PCL performed before a week could be one of the treatment options for acute PCL injury.

• Journal of Korean Orthopaedic Sports Medicine
• /
• v.10 no.1
• /
• pp.18-26
• /
• 2011

Purpose: This study is to compare the clinical results of ACL reconstruction between three groups using hamstring tendon autograft, mixed and tibialis tendon allograft. Materials and Methods: Between August 2003 and August 2008, we analyzed 169 cases of ACL reconstruction, 66 cases used hamstring tendon autograft, 42 cases used mixed graft and 61 cases used tibialis tendon allograft, with a minimum follow-up of 12 months. For the clinical evaluation, we evaluated the Lysholm score, Telos stress test device and IKDC score. Results: The average side to side difference in Telos stress test decreased from $7.5{\pm}1.0$ mm to $1.6{\pm}1.0$ mm in autograft group, from $7.6{\pm}1.1$ mm to $1.4{\pm}1.1$ mm in mixed graft group and from $7.4{\pm}1.3$ mm to $2.5{\pm}1.3$ mm in allograft group. The average Lysholm knee score improved from 58.6 to 92.3 in autograft group, from 60.6 to 92.6 in mixed graft group and from 55.3 to 91.5 in allograft group. There was no significant difference between three groups in clinical results. At second look arthroscopy, tension of ligament and synovial coverage were good result in autograft and mixed graft than allograft group. Conclusion: All hamstring tendon autograft, mixed graft and tibialis tendon allograft groups showed satisfactory clinical results, with no significant difference in outcomes between the groups. Both hamstring tendon autograft and mixed graft showed good synovial coverage in second look arthroscopy. So mixed graft will be considered as good alternative in case of shorter or thin harvested hamstring tendon.