No abstract

Disease specific patterns of volatile organic compounds can be detected in exhaled breath using an electronic nose (e-nose). The aim of this study is to explore whether an e-nose can differentiate between head and neck, and lung carcinoma. Eighty-seven patients received an e-nose measurement before any oncologic treatment. We used PARAFAC/TUCKER3 tensor decomposition for data reduction and an artificial neural network for analysis to obtain binary results; either diagnosed as head and neck or lung carcinoma. Via a leave-one-out method, cross-validation of the data was performed. In differentiating head and neck from lung carcinoma patients, a diagnostic accuracy of 93 % was found. After cross-validation of the data, this resulted in a diagnostic accuracy of 85 %. There seems to be a potential for e-nose as a diagnostic tool in HNC and lung carcinoma. With a fair diagnostic accuracy, an e-nose can differentiate between the two tumor entities.Electronic supplementary materialThe online version of this article (doi:10.1007/s00405-016-4038-x) contains supplementary material, which is available to authorized users.

Electronic nose (e-nose) technology has the potential to detect cancer at an early stage and can differentiate between cancer origins. Our objective was to compare patients who had head and neck squamous cell carcinoma (HNSCC) with patients who had colon or bladder cancer to determine the distinctive diagnostic characteristics of the e-nose. Feasibility study An e-nose device was used to collect samples of exhaled breath from patients who had HNSCC and those who had bladder or colon cancer, after which the samples were analyzed and compared. One hundred patients with HNSCC, 40 patients with bladder cancer, and 28 patients with colon cancer exhaled through an e-nose for 5 min. An artificial neural network was used for the analysis, and double cross-validation to validate the model. In differentiating HNSCC from colon cancer, a diagnostic accuracy of 81 % was found. When comparing HNSCC with bladder cancer, the diagnostic accuracy was 84 %. A diagnostic accuracy of 84 % was found between bladder cancer and colon cancer. The e-nose technique using double cross-validation is able to discriminate between HNSCC and colon cancer and between HNSCC and bladder cancer. Furthermore, the e-nose technique can distinguish colon cancer from bladder cancer.

2b.

Background The aim of this feasibility study was to assess the diagnostic performance of an electronic nose (e‐nose) as a noninvasive diagnostic tool in detecting locoregional recurrent and/or second (or third) primary head and neck squamous cell carcinoma (HNSCC) after curative treatment. Methods Using an e‐nose (Aeonose, The eNose Company, Zutphen, The Netherlands), breath samples were collected from patients after curative treatment of an HNSCC with a locoregional recurrence or second (or third) primary tumor (N = 20) and from patients without evidence of recurrent disease (N = 20). Analyses were performed utilizing artificial neural networking based on patterns of volatile organic compounds. Results A diagnostic accuracy of 83% was observed in differentiating follow‐up patients with locoregional recurrent or second (or third) primary HNSCC from those without evidence of disease. Conclusion This study has demonstrated the feasibility of using an e‐nose to detect locoregional recurrent and/or second (or third) primary HNSCC.

Introduction Detecting volatile organic compounds in exhaled breath enables the diagnosis of cancer. We investigated whether a handheld version of an electronic nose is able to discriminate between patients with head and neck squamous cell cancer (HNSCC) and healthy controls. Methods Ninety‐one patients with HNSCC and 72 controls exhaled through an e‐nose. An artificial neural network based model was built to separate between HNSCC patients and healthy controls. Additionally, three models were created for separating between the oral, oropharyngeal, and glottic subsites respectively, and healthy controls. Results The results showed a diagnostic accuracy of 72% at a sensitivity of 79%, specificity of 63%, and area under the curve (AUC) of 0.75. Results for the subsites showed an AUC of 0.85, 0.82, and 0.83 respectively for oral, oropharyngeal, and glottic HNSCC. Conclusion This feasibility study showed that this portable noninvasive diagnostic tool can differentiate between HNSCC patients and healthy controls.

This study determines the relationship between patient and investigator reported outcome measures (PROMs versus IROMs) on oropharyngeal dysphagia (OD) in Parkinson’s disease (PD). The PROMs used are the MD Anderson Dysphagia Inventory (MDADI) and the Dysphagia Severity Scale (DSS). The IROMs used are fiberoptic endoscopic evaluation of swallowing (FEES) and videofluoroscopy of swallowing (VFS). Ninety dysphagic PD patients were included. Multilayer perceptron (MLP) neural network analysis was used to investigate the relationship between PROMs and IROMs on OD in PD. MLP neural network analysis showed a moderate agreement between PROMs and IROMs, with an area under the curve between 0.6 and 0.7. Two-step cluster analysis revealed several clusters of patients with similar scores on FEES and/or VFS variables, but with significant different scores on MDADI and DSS variables. This study highlights that there are PD patients with similar FEES and/or VFS findings that cannot be lumped together under the same pathophysiological umbrella due to their differences in PROMs. Since the exact origin of these differences is not fully understood, it seems appropriate for the time being to take into account the different dimensions of OD during the swallowing assessment so that they can be included in a patient-tailored treatment plan.