Deep Upscale U-Net for automatic tongue segmentation.

In a treatment or diagnosis related to oral health conditions such as oral cancer and oropharyngeal cancer, an investigation of tongue's movements is a major part. In an automatic measurement of such movement, it must first start with a task of tongue segmentation. This paper proposes a solution of tongue segmentation based on a decoder-encoder CNN-based structure i.e., U-Net. However, it could suffer from a problem of feature loss in deep layers. This paper proposes a Deep Upscale U-Net (DU-UNET). An additional up-sampling of the feature map from a contracting path is concatenated to an upper layer of an expansive path, based on an original U-Net structure. The segmentation model is constructed by training DU-UNET on the two publicly available datasets, and transferred to the self-collected dataset of tongue images with five tongue postures which were recorded at a far distance from a camera under a real-world scenario. The proposed DU-UNET outperforms the other existing methods in our literature reviews, with accuracy of 99.2%, mean IoU of 97.8%, Dice score of 96.8%, and Jaccard score of 96.8%.

Encoder-decoder network with RMP for tongue segmentation.

Tongue and its movements can be used for several medical-related tasks, such as identifying a disease and tracking a rehabilitation. To be able to focus on a tongue region, the tongue segmentation is needed to compute a region of interest for a further analysis. This paper proposes an encoder-decoder CNN-based architecture for segmenting a tongue in an image. The encoder module is mainly used for the tongue feature extraction, while the decoder module is used to reconstruct a segmented tongue from the extracted features based on training images. In addition, the residual multi-kernel pooling (RMP) is also applied into the proposed network to help in encoding multiple scales of the features. The proposed method is evaluated on two publicly available datasets under a scenario of front view and one tongue posture. It is then tested on a newly collected dataset of five tongue postures. The reported performances show that the proposed method outperforms existing methods in the literature. In addition, the re-training process could improve applying the trained model on unseen dataset, which would be a necessary step of applying the trained model on the real-world scenario.

Effect of TRP-Stimulating Compounds to Reduce Swallowing Response Time in the Elderly: A Systematic Review.

There is still controversy on evidence supporting compounds that stimulate the transient receptor potential cation channel (TRP) receptor, which can reduce the latency of swallowing reflex. This systematic review aims to evaluate the effectiveness of TRP-stimulating compounds to reduce swallowing reflex time in the elderly. We searched the following databases: PubMed, EMBASE, OVID, Central, Scopus, ISI, CINALH, LILACS, CRD database, and Open grey until June 1st, 2019. We included randomized controlled trials (RCTs) which compared swallowing time between TRP-stimulating compounds and placebo or no treatment in population aged more than 60 years old. We assessed bias using the Cochrane risk of bias tool. Three authors independently screened and selected studies. Quality assessment and data extraction were performed by two authors independently. Of 363 reviews, we included four RCTs. Two RCTs used black pepper, one RCT used capsaicinoids, and the other one used capsaicin ointment to stimulate the TRPV1 receptor. All of the studies were assessed at unclear bias except the study, which used capsaicin ointment with assessed as low risk of bias. Meta-analysis could not be done in this study due to the different baseline characteristics and definition of swallowing reflex time. This review demonstrates the potential effect of TRPV1-stimulating compounds to reduce swallowing response time in the elderly were remains unclear. Most studies had an unclear bias. Further larger and well-designed RCTs are needed to draw robust conclusions.

The effect of reclining position on swallowing function in stroke patients with dysphagia.

BACKGROUND: Dysphagia is a common problem in patients with a history of stroke. In Japan, a reclined position is commonly used as a compensatory technique to address this problem. OBJECTIVE: To evaluate the effect of reclined position on swallowing function in patients with stroke who had dysphagia. METHODS: A retrospective analysis was carried out on the videofluoroscopic examination of swallowing (VF) of 4ml honey-thick liquid swallows collected over 9 years. Penetration-aspiration scale (PAS) and residue scores were compared for the following: a body position at 90° upright (90°U) and 60° reclining (60°R) groups, as well as 60°R and 45° reclining (45°R) groups. RESULTS: Two hundred and five records from 98 subjects were reviewed. These included patients with ischaemic stroke (62%), haemorrhagic stroke (32%) and subarachnoid haemorrhage (6%). PAS scores were lower when the body was in a more reclined position (P < .001). The amount of residue in the valleculae and pyriform sinus also reduced in the more reclined position (P < .001). The deeper bolus head at swallowing onset was positively correlated with severe PAS (P < .001). CONCLUSIONS: These findings suggest that in patients with stroke who had dysphagia, a reclined position may be useful in reducing the risk of penetration and aspiration, and in decreasing the amount of residue in the pharyngeal area. The depth of the bolus head at the onset of swallowing increases the severity of penetration and aspiration.

The prevalence and findings of fibre-optic endoscopic evaluation of swallowing in hospitalised patients with dysphagia.

Swallowing disorder or dysphagia is quite common in hospitalised patients. Using fibre-optic endoscopic evaluation of swallowing (FEES) is one of the clinical standards for evaluating swallowing disorder to prevent serious consequences such as aspiration pneumonia. This study aimed to determine the prevalence and the associated risk of dysphagia in hospitalised patients by using FEES finding. We retrospectively analysed the FEES records from the patients who were screened and suspected of swallowing problems by a certified nurse of dysphagia nursing (CNDN). The FEES findings were compared between dysphagia and without dysphagia to evaluate the associated risk of dysphagia. Six-hundred and nine FEES records were analysed. We found dysphagia 76% in patients who suspected swallowing problems by CNDN. FEES was assessed after the subjects had been admitted for 22 days on average. There was no difference in age between dysphagia and without dysphagia participants. However, the advanced age (age > 85 years old) increased the odd of dysphagia 1.18, P = .03. The primary disease of the subjects was mainly cerebrovascular disease (24%) and pneumonia (22%). Abnormal FEES findings including soft palate elevation, velopharyngeal contraction, whiteout, volitional cough, glottis closure during breath holding, cough reflex and presence of secretion in pharynx were found in hospitalised patients with dysphagia. The prevalence of dysphagia was high in hospitalised patients. Hence, screening the swallowing problem by nurse and FEES evaluation is essential to detect and prevent the complication in the patient who has dysphagia.