Cervical Spondylosis Diagnosis Based on Convolutional Neural Network with X-ray Images.

The increase in Cervical Spondylosis cases and the expansion of the affected demographic to younger patients have escalated the demand for X-ray screening. Challenges include variability in imaging technology, differences in equipment specifications, and the diverse experience levels of clinicians, which collectively hinder diagnostic accuracy. In response, a deep learning approach utilizing a ResNet-34 convolutional neural network has been developed. This model, trained on a comprehensive dataset of 1235 cervical spine X-ray images representing a wide range of projection angles, aims to mitigate these issues by providing a robust tool for diagnosis. Validation of the model was performed on an independent set of 136 X-ray images, also varied in projection angles, to ensure its efficacy across diverse clinical scenarios. The model achieved a classification accuracy of 89.7%, significantly outperforming the traditional manual diagnostic approach, which has an accuracy of 68.3%. This advancement demonstrates the viability of deep learning models to not only complement but enhance the diagnostic capabilities of clinicians in identifying Cervical Spondylosis, offering a promising avenue for improving diagnostic accuracy and efficiency in clinical settings.

A Coherent Wideband Acoustic Source Localization Using a Uniform Circular Array.

In modern applications such as robotics, autonomous vehicles, and speaker localization, the computational power for sound source localization applications can be limited when other functionalities get more complex. In such application fields, there is a need to maintain high localization accuracy for several sound sources while reducing computational complexity. The array manifold interpolation (AMI) method applied with the Multiple Signal Classification (MUSIC) algorithm enables sound source localization of multiple sources with high accuracy. However, the computational complexity has so far been relatively high. This paper presents a modified AMI for uniform circular array (UCA) that offers reduced computational complexity compared to the original AMI. The complexity reduction is based on the proposed UCA-specific focusing matrix which eliminates the calculation of the Bessel function. The simulation comparison is done with the existing methods of iMUSIC, the Weighted Squared Test of Orthogonality of Projected Subspaces (WS-TOPS), and the original AMI. The experiment result under different scenarios shows that the proposed algorithm outperforms the original AMI method in terms of estimation accuracy and up to a 30% reduction in computation time. An advantage offered by this proposed method is the ability to implement wideband array processing on low-end microprocessors.

A Lower Bound on the Estimation Variance of Direction-of-Arrival and Skew Angle of a Biaxial Velocity Sensor Suffering from Stochastic Loss of Perpendicularity.

The biaxial velocity sensor comprises two nominally perpendicular particle velocity sensors and a collocated pressure sensor. Due to real-world imperfections in manufacturing or setup errors, the two axes may suffer from perpendicularity losses. To analytically study how skewness affects its direction-finding performance, the hybrid Cramér-Rao bound (HCRB) of the directions-of-arrival for the polar angle, azimuth angle and the skew angle of a biaxial velocity sensor that suffers from stochastic loss of perpendicularity were derived in closed form. The skew angle was modeled as a zero-mean Gaussian random variable of a known variance, which was assumed to be very small, to capture the uncertainty in the orthogonality of the biaxial velocity sensor. The analysis shows that for the polar and azimuth angle, the loss of perpendicularity introduces the variation of the HCRB along the azimuth angle axis, which is independent of the skew angle, but on its variance. The dynamic range of this variation increases as the variance of the skew angle increases. For the estimation of the skew angle, the HCRB of the skew angle is bounded upwards by the variance of the skew angle and varies with the azimuth angle. The hybrid maximum likelihood- maximum a posterior (hybrid ML/MAP) estimator was used to verify the derived bounds.

A deep learning approach for detecting drill bit failures from a small sound dataset.

Monitoring the conditions of machines is vital in the manufacturing industry. Early detection of faulty components in machines for stopping and repairing the failed components can minimize the downtime of the machine. In this article, we present a method for detecting failures in drill machines using drill sounds in Valmet AB, a company in Sundsvall, Sweden that supplies equipment and processes for the production of pulp, paper, and biofuels. The drill dataset includes two classes: anomalous sounds and normal sounds. Detecting drill failure effectively remains a challenge due to the following reasons. The waveform of drill sound is complex and short for detection. Furthermore, in realistic soundscapes, both sounds and noise exist simultaneously. Besides, the balanced dataset is small to apply state-of-the-art deep learning techniques. Due to these aforementioned difficulties, sound augmentation methods were applied to increase the number of sounds in the dataset. In this study, a convolutional neural network (CNN) was combined with a long-short-term memory (LSTM) to extract features from log-Mel spectrograms and to learn global representations of two classes. A leaky rectified linear unit (Leaky ReLU) was utilized as the activation function for the proposed CNN instead of the ReLU. Moreover, an attention mechanism was deployed at the frame level after the LSTM layer to pay attention to the anomaly in sounds. As a result, the proposed method reached an overall accuracy of 92.62% to classify two classes of machine sounds on Valmet's dataset. In addition, an extensive experiment on another drilling dataset with short sounds yielded 97.47% accuracy. With multiple classes and long-duration sounds, an experiment utilizing the publicly available UrbanSound8K dataset obtains 91.45%. Extensive experiments on our dataset as well as publicly available datasets confirm the efficacy and robustness of our proposed method. For reproducing and deploying the proposed system, an open-source repository is publicly available at https://github.com/thanhtran1965/DrillFailureDetection_SciRep2022 .

A Study on the Correlation between Change in the Geometrical Dimension of a Free-Falling Molten Glass Gob and Its Viscosity.

To produce flawless glass containers, continuous monitoring of the glass gob is required. It is essential to ensure production of molten glass gobs with the right shape, temperature, viscosity and weight. At present, manual monitoring is common practice in the glass container industry, which heavily depends on previous experience, operator knowledge and trial and error. This results in inconsistent measurements and consequently loss of production. In this article, a multi-camera based setup is used as a non-invasive real-time monitoring system. We have shown that under certain conditions, such as keeping the glass composition constant, it is possible to do in-line measurement of viscosity using sensor fusion to correlate the rate of geometrical change in the gob and its temperature. The correlation models presented in this article show that there is a strong correlation, i.e., 0.65, between our measurements and the projected viscosity.

A Deep CNN Transformer Hybrid Model for Skin Lesion Classification of Dermoscopic Images Using Focal Loss.

Skin cancers are the most cancers diagnosed worldwide, with an estimated > 1.5 million new cases in 2020. Use of computer-aided diagnosis (CAD) systems for early detection and classification of skin lesions helps reduce skin cancer mortality rates. Inspired by the success of the transformer network in natural language processing (NLP) and the deep convolutional neural network (DCNN) in computer vision, we propose an end-to-end CNN transformer hybrid model with a focal loss (FL) function to classify skin lesion images. First, the CNN extracts low-level, local feature maps from the dermoscopic images. In the second stage, the vision transformer (ViT) globally models these features, then extracts abstract and high-level semantic information, and finally sends this to the multi-layer perceptron (MLP) head for classification. Based on an evaluation of three different loss functions, the FL-based algorithm is aimed to improve the extreme class imbalance that exists in the International Skin Imaging Collaboration (ISIC) 2018 dataset. The experimental analysis demonstrates that impressive results of skin lesion classification are achieved by employing the hybrid model and FL strategy, which shows significantly high performance and outperforms the existing work.

Multi-Camera Based Setup for Geometrical Measurement of Free-Falling Molten Glass Gob.

High temperatures complicate the direct measurements needed for continuous characterization of the properties of molten materials such as glass. However, the assumption that geometrical changes when the molten material is in free-fall can be correlated with material characteristics such as viscosity opens the door to a highly accurate contactless method characterizing small dynamic changes. This paper proposes multi-camera setup to achieve accuracy close to the segmentation error associated with the resolution of the images. The experimental setup presented shows that the geometrical parameters can be characterized dynamically through the whole free-fall process at a frame rate of 600 frames per second. The results achieved show the proposed multi-camera setup is suitable for estimating the length of free-falling molten objects.

Paperboard Coating Detection Based on Full-Stokes Imaging Polarimetry.

The manufacturing of high-quality extruded low-density polyethylene (PE) paperboard intended for the food packaging industry relies on manual, intrusive, and destructive off-line inspection by the process operators to assess the overall quality and functionality of the product. Defects such as cracks, pinholes, and local thickness variations in the coating can occur at any location in the reel, affecting the sealable property of the product. To detect these defects locally, imaging systems must discriminate between the substrate and the coating. We propose an active full-Stokes imaging polarimetry for the classification of the PE-coated paperboard and its substrate (before applying the PE coating) from industrially manufactured samples. The optical system is based on vertically polarized illumination and a novel full-Stokes imaging polarimetry camera system. From the various parameters obtained by polarimetry measurements, we propose implementing feature selection based on the distance correlation statistical method and, subsequently, the implementation of a support vector machine algorithm that uses a nonlinear Gaussian kernel function. Our implementation achieves 99.74% classification accuracy. An imaging polarimetry system with high spatial resolution and pixel-wise metrological characteristics to provide polarization information, capable of material classification, can be used for in-process control of manufacturing coated paperboard.

Single-Nucleotide Polymorphisms and Cancer Risk, Tumor Recurrence, or Survival of Head and Neck Cancer Patients.

OBJECTIVE: This paper aims at studying the influence of single-nucleotide polymorphisms (SNPs) on cancer risk, tumor recurrence, and survival in head and neck (H&N) cancer patients. METHODS: A total of 45 SNPs in 41 genes were investigated. A total of 174 Caucasian H&N cancer patients and 245 healthy blood donors were enrolled in the study. RESULTS: Ten SNPs were associated with H&N cancer risk, but the identified SNPs differed among males and females. Some of the SNPs were related to immune response genes. The immune response gene SNPs were also related to survival. In particular, we noted that the tumor necrosis factor alpha (TNFα) rs1800629 could have an influence on cancer risk, tumor recurrence as well as survival. CONCLUSION: Genetic variation of the TNFα rs1800629 might be useful as a biomarker in clinical decision-making since it was found to be related to cancer risk, tumor recurrence, and survival of H&N cancer patients.

Perforin, CD28 and CD95 expression in circulating CD4 and CD8 cells as predictors of head and neck (H&N) cancer patient survival.

Long-term survival of H&N cancer patients has not improved significantly over the last 30 years. The possibility of using circulating blood cell phenotypes as a prognostic biomarker of H&N cancer patient was investigated in this study. Pre-treatment, circulating T lymphocyte subpopulations as well as the survival time of the patients in question were studied. Upregulated CD4+ perforin+ and CD8+ CD95+ but downregulated CD4+ CD28+ (p < 0.001) were detected in H&N cancer patients. With 3 years of follow-up time, an increase in the frequency of the pre-treatment, circulating CD4+ perforin+ cells and CD8+ perforin+ cells was showed to have reverse effects on the survival time in H&N cancer patients (p < 0.01). Detection of perforin+ frequency in CD4+ and CD8+ lymphocyte by FACS is fast, simple and cost-effective. A potential role of perforin expression in CD4+ and CD8+ cells as a prognostic biomarker for H&N cancer patient in the clinical setting was suggested.

Plasma tumor necrosis factor-α and C-reactive protein as biomarker for survival in head and neck squamous cell carcinoma.

PURPOSE: Tumor TNM staging is the main basis for prognosis and treatment decision for head and neck squamous cell carcinoma (HNSCC) despite significant heterogeneity in terms of outcome among patients with the same clinical stage. In this study, a possible role of plasma interleukin-2 (IL-2), interleukin-6 (IL-6), granulocyte-macrophage colony-stimulating factor (GM-CSF), tumor necrosis factor-α (TNF-α) and C-reactive protein (CRP) as biomarkers for survival of HNSCC patients was investigated. METHODS: In this prospective study, plasma levels of IL-2, IL-6, GM-CSF, TNF-α and CRP in patients (n = 100) and controls (n = 48) were analyzed. RESULTS: Significantly elevated levels of CRP and TNF-α (p < 0.001) were found in the patients. Combination of upregulated CRP and TNF-α in the patient plasma was significantly related to shorter patient survival, independent of clinical stage. CONCLUSIONS: Our findings indicate that CRP and TNF-α might be suitable as biomarkers in combination with tumor TNM staging for predicting survival and individualized treatment of HNSCC patients. Plasma CRP and TNF-α analysis are simple, rapid, cost effective and suitable for clinical practice.

DNA content and methylation of p16, DAPK and RASSF1A gene in tumour and distant, normal mucosal tissue of head and neck squamous cell carcinoma patients.

UNLABELLED: Long-term survival of head and neck squamous cell carcinoma (HNSCC) patients has not improved significantly during the last 20 years and recurrent disease is frequently observed. In this study, the potential presence of pre-malignant cells or rare malignant cells at the time of diagnosis in HNSCC was investigated. PATIENTS AND METHODS: Fifty-nine biopsies obtained from 41 HNSCC patients were analysed. Eighteen of these biopsies were normal mucosal tissue, located at least 5 cm from the tumour margin. DNA content and DNA methylation of p16, DAPK and RASSF1A was examined. RESULTS: Thirty-nine out of 41 (95%) tumour biopsies showed p16 methylation and 21 (51%) of them displayed aneuploidy. Of 18 distant normal mucosal biopsies, 6 (33%) of these showed evidence of aneuploidy and 15(83%) of them showed methylated p16 genes. Among paired samples, the highest frequencies of DNA methylation were found in tumours with aneuploidy. Regardless of DNA content, methylation at DAPK, RASSF1A or p16 were found in the corresponding distant mucosal biopsies. CONCLUSION: The cells with abnormal DNA content or DNA methylation in mucosal tissue were not detected clinically or by pathological macroscopic and microscopic examination. Thus, distant mucosal tissue DNA content and DNA methylation analyses in combination with histopathology will provide a better prognostic base for the evaluation and treatment of HNSCC patients.

Hypoglossal schwannoma-successful reinnervation and functional recovery of the tongue following tumour removal and nerve grafting.

OBJECTIVE: Hypoglossal nerve schwannomas are rare tumours that usually cause ipsilateral hypoglossal palsy. This report describes such lesions in two patients and suggests nerve grafting as part of the treatment regimen. METHOD: Two patients with intra- and extra-dural hypoglossal schwannomas respectively were treated by direct surgery via a postero-lateral approach to the posterior fossa, hypoglossal canal and carotid sheath. Following tumour removal, sural nerve grafting was used to reconstruct the nerves. Unexpectedly, muscle bulk and motor function returned within 6 months in both patients. CONCLUSION: Nerve grafting was highly successful in achieving functional recovery following surgery for hypoglossal nerve schwannomas.

In vitro effect of radiation, antibody to epidermal growth factor receptor and Docetaxel in human head and neck squamous carcinoma cells with mutant P53 and over-expressed EGFR.

PURPOSE: Radiotherapy is the most frequently used and cheapest treatment both for curative and palliative purposes in HNSCC. Despite advances in technology and intensive treatments with radiation, only half of the patients are cured. New therapeutic approaches focusing on the molecular mechanism that mediate tumour cell growth or cell death in combination with radiotherapy have been suggested. The effects of radiation, antibody to EGFR and Docetaxel as single treatment or in combinations on HNSCC cells were investigated. METHODS: The established HNSCC cells with mutant (mt) P53 and over-expressed normal EGFR was used as the in vitro model. Gene expression profile, cell cycle progression and cell death were used as the indication of treatment outcome. RESULTS: With c-DNA microarray of well-characterised functional genes, massive changes in the genes expression of HNSCC were detected. The alterations of gene expression profiles do not have any correlation neither on tumour cell growth nor cell death. HNSCC cells with mt P53 and over-expressed normal EGFR did not response to radiation, anti-EGFR monoclonal antibody and their combination therapy. Effective treatment could be obtained from single therapy with Docetaxel. No additive effects on cell cycle arrest or cell death were seen in the combination of Docetaxel to anti-EGFR antibody, radiation or anti-EGFR antibody + radiation. CONCLUSIONS: The c-DNA microarray analysis does not indicate any specific target or treatment effects of HNSCC with mt P53 and over-expressed normal EGFR. Single therapy, target at microtubules might be the most suitable treatment modulation in this tumour type.

Clinical outcome following radiotherapy and planned neck dissection in N+ head and neck cancer patients.

CONCLUSIONS: This study confirms earlier findings that patients with viable tumour cells in the neck after external beam radiotherapy (EBRT) have a poor prognosis. The study also indicates that neck dissection (ND) does not change the prognosis for patients with a complete clinical response in the neck. At the moment our guidelines concerning this matter are being reviewed. OBJECTIVES: The protocol at our institution stipulates a planned ND in patients with metastasis in the neck after EBRT regardless of the response in the neck. As the necessity for a planned ND has not been clarified we wanted to evaluate our results. PATIENTS AND METHODS: Patients diagnosed from 1998 to 2002 with metastasis in the neck who received EBRT were evaluated for histopathological findings and clinical outcome. RESULTS: A total of 156 patients were included. Overall survival was 62% and disease-specific survival was 76%. There was a complete response (CR) in the neck in 63 patients (40%); among these 15 had viable tumour cells in the neck. In patients not achieving CR, 40% (37/93) had viable tumour cells left in the neck. Patients with viable tumour cells in the neck after EBRT had disease-specific survival of 48% compared with 90% among patients without viable tumour cells.

Mobile phone use and risk of parotid gland tumor.

Handheld mobile phones were introduced in Denmark and Sweden during the late 1980s. This makes the Danish and Swedish populations suitable for a study aimed at testing the hypothesis that long-term mobile phone use increases the risk of parotid gland tumors. In this population-based case-control study, the authors identified all cases aged 20-69 years diagnosed with parotid gland tumor during 2000-2002 in Denmark and certain parts of Sweden. Controls were randomly selected from the study population base. Detailed information about mobile phone use was collected from 60 cases of malignant parotid gland tumors (85% response rate), 112 benign pleomorphic adenomas (88% response rate), and 681 controls (70% response rate). For regular mobile phone use, regardless of duration, the risk estimates for malignant and benign tumors were 0.7 (95% confidence interval: 0.4, 1.3) and 0.9 (95% confidence interval: 0.5, 1.5), respectively. Similar results were found for more than 10 years' duration of mobile phone use. The risk estimate did not increase, regardless of type of phone and amount of use. The authors conclude that the data do not support the hypothesis that mobile phone use is related to an increased risk of parotid gland tumors.

Significance of RNA reference in tumour-related gene expression analyses by cDNA array.

The cDNA array technique is an efficient approach for studying the expression of a large number of genes in a single experiment. The cDNA array analysis indicates the relative level of corresponding gene expression from a specimen and a reference. Our investigation was performed to address the significance of reference RNA on the outcome of the cancer-related gene expression profile obtained from cDNA array analysis. Human head and neck squamous cell carcinoma (HNSCC) biopsies and 5 sources of RNA reference were used for this purpose. In these biopsies, each individual patient expressed a unique set of genes both in normal and tumour tissue. It is important to note that 5 striking patterns of tumour-related gene expression were obtained according to the 5 references used. Significant differences in 60%, 16%, 15% and 15% of the genes expressed were shown when autologous normal matched tissue biopsy references were compared to pooled cell lines, allogenic normal mixed cell types, tumours or allogenic normal matched cell type references, respectively. Thus, theoretically and our study suggested that patient autologous normal cells matching with the tumour type should be the most suitable reference in cDNA array for the identification of individual tumour gene profiles with clinical purpose.

Telehomecare: quality, perception, satisfaction.

The aim of this study was to demonstrate that telehomecare linking homebound patients with their home health-care nurses over the plain old telephone system (POTS) provides high-quality, clinically useful, and patient satisfactory interactions. Congestive heart failure, chronic obstructive pulmonary disease, and chronic wound-care patients receiving skilled home nursing care were randomized into control (standard home health care, HHC) and two intervention (standard care plus video conferencing/Internet access; the above plus physiological monitoring) groups. Virtual visits (VVs), consisting of two-way audio and video interactions between the central site HHC nurse and the subject at home, were compared for technical quality and clinical usefulness by the HHC nurses who performed the VVs. Subject perception of telehomecare and satisfaction with their HHC were assessed over the course of the project. There were a total of 567 virtual and 1,057 actual visits conducted for the 53 subjects completing the study. The technical quality of VVs were rated at 94.7%. They were considered to be as useful as actual visits in 90.7% of cases. Subject telehomecare perception increased after experiencing the process. All subjects were satisfied with their HHC; satisfaction increased with an increasing level of telehomecare intervention. Subjects receiving physiological monitoring and video conferencing/Internet access in addition to standard care were most satisfied with their care. VVs can be conducted over POTS. Patients can use telehomecare with moderate levels of training. These programs can provide timely and quality home health nursing care with VVs augmenting traditional home visits.

When does my treatment start?--The continuum of care for patients with head and neck cancer.

BACKGROUND AND PURPOSE: The clinical workup of patients with head and neck (H&N) cancer is often time-consuming. The overall time involved is obviously important to the treatment results, since delays in time can increase the risk for clinical upstaging resulting in a worse outcome. Although for ethical considerations, time delay is impossible to study in randomised trials. Major reorganisations for this group of patients have taken place in Stockholm during the 1990s. In this study, we measured the duration of the overall continuum of care experienced by the patients, in different time periods during the 1990s. MATERIALS AND METHODS: Two hundred and two patients from the population base of southern Stockholm were identified during different time periods. The continuum of care was identified as the period between the first presentation of the patient to the health care system to the first date of the patient's treatment for his/her cancer. This period was divided into several intervals reflecting the patient's perspective of his or her continuum of care. These intervals were then compared. RESULTS: Median time from first consultation to start of treatment increased between 1994 and 1999 from 67 to 89 days (P = 0.018). The increase of time occurred from referral from first care provider to ENT-specialist, from first visit to ENT-specialist to date of diagnosis. CONCLUSIONS: Different parameters such as duration of the continuum of care must be monitored before and after reorganisations.