We consider PET (positron emission tomography) measurement with SPM (Statistical Parametric Mapping) analysis to be one of the most useful methods to identify activated areas of the brain involved in language processing. SPM is an effective analytical method that detects markedly activated areas over the whole brain. However, with the conventional presentations of these functional brain images, such as horizontal slices, three directional projection, or brain surface coloring, makes understanding and interpreting the positional relationships among various brain areas difficult. Therefore, we developed three-dimensionally reconstructed images from these functional brain images to improve the interpretation.
The subjects were 12 normal volunteers. The following three types of images were constructed: 1) routine images by SPM, 2) three-dimensional static images, and 3) three-dimensional dynamic images, after PET images were analyzed by SPM during daily dialog listening. The creation of images of both the three-dimensional static and dynamic types employed the volume rendering method by VTK (The Visualization Toolkit). Since the functional brain images did not include original brain images, we synthesized SPM and MRI brain images by selfmade C++ programs. The three-dimensional dynamic images were made by sequencing static images with available software. Images of both the three-dimensional static and dynamic types were processed by a personal computer system.
Our newly created images showed clearer positional relationships among activated brain areas compared to the conventional method.
To date, functional brain images have been employed in fields such as neurology or neurosurgery, however, these images may be useful even in the field of otorhinolaryngology, to assess hearing and speech. Exact threedimensional images based on functional brain images are important for exact and intuitive interpretation, and may lead to new developments in brain science. Currently, the surface model is the most common method of threedimensional display. However, the volume rendering method may be more effective for imaging regions such as the brain.

View full abstract

Swallowing a corrosive agent (alkali or acid) often causes severe pharyngeal, laryngeal or esophageal stricture (caustic stenosis), which is usually very difficult to treat. This paper reports two cases of esophageal stricture treated by esophagoplasty. Both cases had attempted suicide by swallowing a sodium hydroxide solution or acid.
Case 1 was a 66-year-old man found to have severe hypopharynx and thoracic esophagus stenosis with supraglottic stricture. The supraglottic stricture was reconstructed with an ileocolon graft and laryngectomy. The intestinal anastomosis was patent, but the peristaltic motion in the ileocolon was not good. The patient continues to have difficulty achieving sufficient oral feeding and to receive supplemental feeding via a jejunostomy.
Case 2 was a 81-year-old woman with severe thoracic esophagus stenosis after gastrectomy. The lesion was reconstructed with a jejenum graft. The intestinal anastomosis was patent. She achieved oral alimentation of both liquids and solids without aspiration after surgery.
Esophagectomy in these cases can be difficult and hazardous due to extensive fibrosis and many adhesions to adjacent structures. In both cases, the reconstructed intestine passed through the ante-sternal route, so there was severe scar formation in the mediastinum, and an esophago-skin fistula formed in the cervical skin. Cervical vessels and intestinal vessels were anastomosed for blood supply to the reconstructed intestinal tract. This method is useful because it is safe and results in good deglution.

View full abstract

We used a miniature ultrasonic probe to study its diagnostic application in the dection of laryngeal lesions. Using the immersion method and an extracted human larynx, we evaluated ultrasonic images, identified the layered structure of vocal folds, and examined clinical cases following the study of diagnostic application.
By the immersion method, ultrasonic images revealed that the membranous vocal fold was separated into three layers. We identified the histological structures of these three layers by comparing the ultrasonic images with the corresponding histologies. The first layer (hyperechoic) was the epithelial and the superficial layer of the lamina propria; the second layer (hypoechoic) was the vocal ligament; and the third layer (hyperechoic) was the deep layer of the lamina propria.
On the basis of this study, we then evaluated ultrasonographic images of clinical cases. Benign lesions such as vocal nodules or vocal polyps were noted on the layered structure of the mucosa, especially, polyps as hyperechoic images. Furthermore, glottic cancer lesions demonstrated hypoechoic images and sometimes revealed destruction of the layered structures.
These results suggest that intralaryngeal ultrasonography can be useful as a clinical application for the detection of tumor involvement in the vocal folds.

View full abstract

In advanced oral or mesopharyngeal cancer requiring reconstruction, preventive tracheostomy has been performed to insure an airway postoperatively. However, tracheostomy often causes difficulty in swallowing and stress in the postoperative period for the patients. We have devised a strategy to avoid tracheostomy as much as possible in the cases requiring reconstruction except for the cases with over 2/3 resection of the base of the tongue or with prospective difficulty of re-intubation.
Eleven cases were treated without tracheostomy out of 20 cases operated on with reconstruction in our hospital from 1995 through 1997. These patients were managed under nasal intubation for less than 2 days on average with careful observation postoperatively. In 11 non-tracheostomized cases, 1 case experienced venous thrombosis of a free flap and tracheostomy was performed at the salvage operation. Other 10 non-tracheostomized case were compared with 10 tracheostomized cases who were treated from 1991 though 1995.
It was revealed that non-tracheostomized cases tended to have shorter periods until they could speak or eat, and also have a shorter postoperative admission term. In these cases, we have not experienced serious complications such as aspiration pneumonia or dyspnea, instead, special care in the postoperative period was needed. Thus, certain advanced cases who undergo curative resection followed by reconstruction can be managed without tracheostomy. The indication of tracheostomy in surgery for oral or mesopharyngeal cancer should be decided carefully.

View full abstract

There have been many reports on the synkinesis of mimic muscles following facial palsy, and many efforts have been made to evaluate the degree of synkinesis. Even in normal cases, associated movements of facial muscles can been seen. Empirically, mimic muscles move in association, e. g., when the lips are opened to grin, the lateral portion of the eyebrow moves, or when the mouth is moved as in whistling, tension around the root of the nose subconsciously builds up. Our purpose was to evaluate the movements of facial mimic muscles quantitatively, to analyze them in various facial expressions, and to obtain basic data to the objective evaluation of associated movement in normal subjects. Thirty-seven normal subjects (18 males and 19 females, aged 20 to 40 years) were included in the study. A total of 24 markers were stuck to each subject's face. The threshold image process which revealed only the markers, made it possible to trace and measure the markers on the face. The shifting positions of the markers showed that they moved in accordance with the facial movement from the stationary phase to the maximum stage of movement. These positions were measured and the traces plotted on the coordinate axis. The shifting of the markers were then numerically expressed as trajectory investigation. The facial movements examined in this study consisted of eye closing, forehead wrinkling, whistling, and grinning. We found that, quantitatively, certain groups of mimic muscles might work together to make one facial expression in all the above mentioned facial movements.
In normal individuals, at least 15-20% of other mimic muscle collaboration is needed in addition to main muscles to create a facial expression. This result can serve as a reference for the evaluation of abnormal associated facial movement.

View full abstract