Deep learning for diagnosis of head and neck cancers through radiographic data: a systematic review and meta-analysis.

PURPOSE: This study aims to review deep learning applications for detecting head and neck cancer (HNC) using magnetic resonance imaging (MRI) and radiographic data. METHODS: Through January 2023, a PubMed, Scopus, Embase, Google Scholar, IEEE, and arXiv search were carried out. The inclusion criteria were implementing head and neck medical images (computed tomography (CT), positron emission tomography (PET), MRI, Planar scans, and panoramic X-ray) of human subjects with segmentation, object detection, and classification deep learning models for head and neck cancers. The risk of bias was rated with the quality assessment of diagnostic accuracy studies (QUADAS-2) tool. For the meta-analysis diagnostic odds ratio (DOR) was calculated. Deeks' funnel plot was used to assess publication bias. MIDAS and Metandi packages were used to analyze diagnostic test accuracy in STATA. RESULTS: From 1967 studies, 32 were found eligible after the search and screening procedures. According to the QUADAS-2 tool, 7 included studies had a low risk of bias for all domains. According to the results of all included studies, the accuracy varied from 82.6 to 100%. Additionally, specificity ranged from 66.6 to 90.1%, sensitivity from 74 to 99.68%. Fourteen studies that provided sufficient data were included for meta-analysis. The pooled sensitivity was 90% (95% CI 0.820.94), and the pooled specificity was 92% (CI 95% 0.87-0.96). The DORs were 103 (27-251). Publication bias was not detected based on the p-value of 0.75 in the meta-analysis. CONCLUSION: With a head and neck screening deep learning model, detectable screening processes can be enhanced with high specificity and sensitivity.

Acceptability of fully guided virtual implant planning software among dental undergraduate students.

BACKGROUND: Fully guided implant surgery as a technique is gaining popularity. It has been observed that use of surgical guides improves precision and predictability for dental implant placement. However, like any other newer technology, the acceptance of fully guided dental implant technology among users is based upon its perceived usability. This study aimed at evaluating the perception about using Virtual Implant Planning Software (VIPS) among undergraduate dental students at the university of Sharjah. METHODS: Ninety-Six dental surgery students from the University of Sharjah were included in the study. One week after the Virtual Implant Planning Software (Planmeca Romexis version 6.2 procedure, students were asked to complete a Combined technology acceptance model and the theory of planned behaviour (C-TAM TPB) questionnaire. Sixty-six students responded to the questionnaire. RESULTS: Cronbach's alpha surpassed 0.7 for perceived usefulness, perceived ease of use, perceived behavioral control, and subjective norm. Attitude and behavioural intention reported Cronbach's alpha values less than 0.7. Spearman's correlation coefficient was significant for all the constructs. Perceived ease of use explained 49%, 33%, and 42% of the variance of perceived usefulness (R2 = 0.49), attitude (R2 = 0.33), and perceived behavioral control (R2 = 0.42) respectively. Perceived usefulness explained 25%, 18%, and 23% of the variance of attitude (R2 = 0.25), behavioral intention (R2 = 0.18), and perceived behavioral control (R2 = 0.23) respectively. Attitude accounted for 25%, 33%, and 29% of the variance of behavioral intention (R2 = 0.25), perceived behavioral control (R = 0.33), and subjective norm (R = 0.29) respectively. CONCLUSION: The fully guided VIPS was acceptable by dental students specifically because of its usability. This makes VIPS a very effective tool for teaching implantology for dental students. VIPS also allows students to perform multiple repetitions of the implant planning procedure which enhances understanding and content retention.

Incidental Findings of Asymptomatic Fungal Infection.

Fungal sinusitis of the paranasal sinuses is a rare infection in healthy individuals but is relatively common in immunocompromised patients. It is often misdiagnosed and frequently a severe disease, as a few forms of it are linked with a higher mortality rate. Effective handling necessitates a speedy analysis and often counts on radiological findings. On cone-beam computed tomography (CBCT) analysis, a bulky polypoid-shaped lesion with a density close to that of soft tissue in CBCT was visualized in the right ethmoid and sphenoid sinuses. There was a significant expansion of the borders of the right ethmoid sinus, and discontinuity or perforation of the sphenoid sinus floor was suspected from CBCT images. Non-contrasted multi-detector computed tomography (MDCT) exhibited opacification and extension of the lesion into the majority of sinuses with dense inspissated materials in the center, which resembled radiographic features of invasive fungal sinusitis. Computed tomography (CT) scan of the maxillofacial region, specifically paranasal sinuses, plays a considerable role in diagnosing fungal sinusitis. In a majority of cases, fungal sinusitis is noticed and diagnosed in immunocompromised patients. However, it is also seen in healthy patients in very rare circumstances, similar to the patient in this report. If the patient is treated rapidly, the prognosis is fair. We present a case of fungal sinusitis in an otherwise healthy young male patient to increase awareness among dental professionals.

Evaluation of streak metal artifacts in cone beam computed tomography by using the Gumbel distribution: a phantom study.

OBJECTIVE: The aim of this study was to confirm whether streak metal artifacts (SMAs) between titanium implants on cone beam computed tomography (CBCT) images could be evaluated by using the Gumbel distribution (GD). Moreover, the influence of different scan settings on SMAs was investigated. STUDY DESIGN: An iodine solution simulating dentin was placed between 2 titanium rods in an acrylic phantom. It was scanned by using CBCT at 2 settings with nearly equivalent exposure doses (90 kV, 7 mA; 78 kV, 10 mA). The images were analyzed, and the dependence of the voxel values in SMAs on GD was investigated with the coefficient of determination (r2). The location parameters, indicating the strength of the SMAs, were calculated for each scan setting and evaluated with the Mann-Whitney U test. Significance was established at p = .05. RESULTS: The SMAs on CBCT images depended on GD (r2 ≥ .959). The SMAs with the 78 kV, 10 mA settings were significantly smaller than those with the 90 kV, 7 mA settings (p < .01). CONCLUSIONS: SMAs on CBCT images could be evaluated by using methods based on GD. The strengths of metal artifacts varied with changes in scan settings, even at nearly equivalent exposure doses.

Cone-beam computed tomographic imaging of silent sinus syndrome: A case series and a literature review.

While silent sinus syndrome (SSS) is familiar to otolaryngologists and ophthalmologists, it is a rare clinical entity in dentistry and is likely to be underdiagnosed due to dentists' lack of awareness of this condition. SSS presents a diagnostic challenge to dentists, as patients typically have no history of trauma or sinusitis. The characteristic feature of SSS is a gradual retreat of the maxillary sinus walls, resulting in enophthalmos and hypoglobus. Multidetector (multislice) computed tomography is the imaging modality of choice for SSS and other paranasal sinus diseases. Cone-beam computed tomography promises to be an alternative low-dose imaging modality. This report describes 3 cases of SSS in adults, who had no identified clinical symptoms except diminutive and opacified maxillary sinuses, as well as the inward bowing of the sinus walls as noted on cone-beam computed tomographic imaging.

Insulin-like growth factor I receptor regulates the radiation-induced G2/M checkpoint in HeLa cells.

Insulin-like growth factor I receptor (IGF-IR) plays pivotal roles in various biological events, including cell growth, transformation, survival, and DNA repair. In this study, we explored its possible involvement in cell cycle checkpoints, using HeLa cells expressing the fluorescent ubiquitination-based cell cycle indicator (Fucci). We found that IGF-IR inhibitor delayed release from radiation-induced G2 arrest, as demonstrated by FACS and pedigree analysis of Fucci fluorescence. Elongated G2 arrest was also induced by inhibitors of phosphatidylinositol-3 kinase (PI3K) and AKT, but not by inhibitor of MEK, which are two major IGF-IR downstream signaling pathways. Double-strand break (DSB) repair kinetics were not affected by IGF-IR inhibitor. CHK1 inhibitor abrogated radiation-induced G2 arrest, whereas radiation-induced phosphorylation of CHK1 at Ser 345 or Ser 296 was decreased by the IGF-IR inhibitor. However, radiation-induced nuclear localization of CHK1 was prolonged in IGF-IR inhibitor-treated cells in comparison with cells that received radiation alone; in the latter, CHK1 returned to the original diffuse distribution in conjunction with release from G2 arrest. We conclude that IGF-IR directly regulates the G2/M checkpoint via the PI3K/AKT pathway without influencing DSB repair, in part by controlling CHK1 localization between the nucleus and cytoplasm.