A Novel Approach for Continuous Health Status Monitoring and Automatic Detection of Infection Incidences in People With Type 1 Diabetes Using Machine Learning Algorithms (Part 2): A Personalized Digital Infectious Disease Detection Mechanism

BACKGROUND

Semisupervised and unsupervised anomaly detection methods have been widely used in various applications to detect anomalous objects from a given data set. Specifically, these methods are popular in the medical domain because of their suitability for applications where there is a lack of a sufficient data set for the other classes. Infection incidence often brings prolonged hyperglycemia and frequent insulin injections in people with type 1 diabetes, which are significant anomalies. Despite these potentials, there have been very few studies that focused on detecting infection incidences in individuals with type 1 diabetes using a dedicated personalized health model.

OBJECTIVE

This study aims to develop a personalized health model that can automatically detect the incidence of infection in people with type 1 diabetes using blood glucose levels and insulin-to-carbohydrate ratio as input variables. The model is expected to detect deviations from the norm because of infection incidences considering elevated blood glucose levels coupled with unusual changes in the insulin-to-carbohydrate ratio.

METHODS

Three groups of one-class classifiers were trained on target data sets (regular days) and tested on a data set containing both the target and the nontarget (infection days). For comparison, two unsupervised models were also tested. The data set consists of high-precision self-recorded data collected from three real subjects with type 1 diabetes incorporating blood glucose, insulin, diet, and events of infection. The models were evaluated on two groups of data: raw and filtered data and compared based on their performance, computational time, and number of samples required.

RESULTS

The one-class classifiers achieved excellent performance. In comparison, the unsupervised models suffered from performance degradation mainly because of the atypical nature of the data. Among the one-class classifiers, the boundary and domain-based method produced a better description of the data. Regarding the computational time, nearest neighbor, support vector data description, and self-organizing map took considerable training time, which typically increased as the sample size increased, and only local outlier factor and connectivity-based outlier factor took considerable testing time.

CONCLUSIONS

We demonstrated the applicability of one-class classifiers and unsupervised models for the detection of infection incidence in people with type 1 diabetes. In this patient group, detecting infection can provide an opportunity to devise tailored services and also to detect potential public health threats. The proposed approaches achieved excellent performance; in particular, the boundary and domain-based method performed better. Among the respective groups, particular models such as one-class support vector machine, K-nearest neighbor, and K-means achieved excellent performance in all the sample sizes and infection cases. Overall, we foresee that the results could encourage researchers to examine beyond the presented features into other additional features of the self-recorded data, for example, continuous glucose monitoring features and physical activity data, on a large scale.

Adaptive Auto-Regressive Proportional Myoelectric Control

In proportional myographic control, one can control either position or velocity of movement. Here, we propose to use adaptive auto-regressive filters, so as to gradually adjust between the two. We implemented this in an adaptive system with closed-loop feedback, where both the user and the machine simultaneously attempt to follow a cursor on a 2-D arena. We tested this on 15 able-bodied and three limb-deficient participants using an eight-channel myoelectric armband. The human-machine pairs learn to perform smoother cursor movements with a larger range of motion when using the auto-regressive filters, as compared with our previous effortswithmoving-average filters. Importantly, the human-machine system converges to an approximate velocity control strategy resulting in faster and more accuratemovements with lessmuscle effort. The method is not specific tomyoelectriccontroland could be used equally well for motion control using high-dimensional signals from reinnervatedmuscles or direct brain recordings.

A fixed point algorithm for extracting the atrial activity in the frequency domain

In this work we present a fixed point algorithm to extract the atrial rhythm in atrial tachyarrhythmias from the surface electrocardiogram (ECG). In the frequency domain the atrial signal is characterized by a concentration of power around a main peak in the bandwidth 3-10Hz. The proposed algorithm exploits this discriminative property of the atrial component in combination with the decoupling of the atrial and other components superposed in the ECG. It recovers only the interesting atrial rhythm in a simple, fast and reliable way using the information contained in all the leads and reducing the average computational time from 0.902s (FastICA) to 0.023s (the proposed method). The algorithm is applied successfully to synthetic and real data. In simulated ECGs, the correlation index obtained was 0.792. In real ECGs, the accuracy of the results was validated using spectral and temporal parameters. The average peak frequency and spectral concentration obtained were 5.354Hz and 59.4%, respectively, and the kurtosis was 0.065.

Constrained temporal extraction of the atrial rhythm in Atrial Fibrillation episodes

The extraction of the Atrial Activity from the Ventricular Activity in Atrial Fibrillation episodes is a must for clinical analysis. We follow the semi Blind Source Extraction S-BSE approach to solve the problem. The proposed algorithm modifies the BSE contrast function to satisfy the prior knowledge about the spectral content of the atrial signal. The introduction of this prior allows obtaining a new algorithm with the following advantages: it allows the extraction of only the atrial component and it improves the quality of the recovered atrial signal in terms of spectral concentration as we show in the results.

Blind detection of nonlinearities in multiple-echo ultrasonic signals

In this paper, we propose and analyze by means of simulations the use of surrogate data algorithms for blind detection of nonlinearities in multiple-echo ultrasonic signals. We assume a blind scheme so that no information about the input (emitted ultrasonic pulse) can be used. The metrics and equations that model some nonlinear situations are carefully reviewed. Also, closed form equations of the third-order metrics from a simplified second-order Volterra kernel are derived. Computer simulations show that the surrogate data technique is a potentially powerful tool for blind detection of nonlinearities in multiple-echo ultrasonic signals if adequate metrics are chosen. They also reveal interesting trade-offs among parameters that model ultrasonic systems and detection percentages.

Clarithromycin with minocycline and clofazimine for Mycobacterium avium intracellulare complex lung disease in patients without the acquired immune deficiency syndrome. GETIM. Groupe d'Etude et de Traitement des Infections à Mycobactéries

SETTING

An open clinical trial for the treatment of Mycobacterium avium intracellulare complex (MAIC) lung disease in human immunodeficiency virus (HIV)-seronegative patients.

OBJECTIVE

To assess the efficacy and tolerance of clarithromycin (0.75-2 g/day) combined with minocycline (200 mg/day) and clofazimine (100 mg/day) for 15 months.

DESIGN

The study was carried out from August 1992 to June 1994 by pulmonologists of various French medical centres. The patients to be enrolled were of either sex, over 18 years of age, HIV-seronegative and suffering from MAIC lung disease, with a confirmed bacteriological and radiological diagnosis. Examinations were to be performed after 1, 2, 3, 6, 9, 12 and 15 months of treatment.

RESULTS

Thirty patients were included, 16 males and 14 females. Eight did not complete the study due to deviations from protocol or adverse effects. The remainder completed the study with a post-treatment follow-up of 27 +/- 7 months. Among the 22 evaluable patients, 18 had a history of lung disease. Tolerance to the drugs was generally good, apart from three cases of hepatic disturbances and three cases of ototoxicity, which required a decrease in clarithromycin dosage after a short interruption of treatment. There were 14 treatment successes, seven treatment failures, defined by absence of bacteriologic conversion, and in one patient the disease evolution remains uncertain.

CONCLUSION

The combination of clarithromycin with minocycline and clofazimine proved effective with persistently negative cultures in 64% of the patients, and an overall good drug tolerance.

First case of mother-to-infant HIV type 1 group O transmission and evolution of C2V3 sequences in the infected child. French HIV Pediatric Cohort Study Group

We report the first case of mother-to-infant transmission and follow-up for an HIV-1 group O virus from Cameroon. Isolates were obtained from the mother at delivery and from the child at birth and when 16 and 30 months old. We analyzed the viral evolution within mother and child by examining 51 sequences spanning C2V3 regions of the viral envelope gene. The mother carried two genotypes, v1 and v2. The genotype v1 was dominant in the child at birth, and persisted as a minor genotype at age 30 months. The genotype v2 was absent in the child sequences. The variability of the nucleotide sequences of the isolates from the child increased with age from 0.8 to 6%, and a novel genotype (v3) appeared at age 30 months. The nonsynonymous-to-synonymous mutation ratio increased with the age of the child, from 0.75 at birth to 1.86 at 30 months, indicating a high rate of fixation of amino acid changes in the child. The overall pattern was similar to that reported by Ganeshan et al. (J Virol 1997;71:663-677) for group M viruses infecting child with a slow-developing form of the disease.

Mucosal T-lymphocytes in central airways of lung transplant recipients

The immunohistochemical profile of mucosal lymphocytes was investigated in the central airways of lung transplant recipients. Bronchial and transbronchial biopsies (BB and TBB, respectively) and bronchoalveolar lavage for culture of bacteria and viruses were performed during a fibroscopic procedure in patients without evidence of chronic rejection, 3 to 10 mo after surgery. Analysis was restricted to samples without concurrent airway infection: 23 pairs of BB and TBB from 18 transplant recipients were analyzed. An immunohistochemical technique was used to identify and score mucosal cells that reacted with monoclonal antibodies against CD4, CD8, CD45-Ro (memory T-cells), and HLA-DR molecules. The same procedure was applied in nine nonsmoking control subjects (NS group). Data from transplant recipients were allocated to R+ (n = 11) or R- groups (n = 12), depending on the presence or absence of histologic evidence of acute rejection on TBB. A statistically significant depletion of every immunoreactive cell subset was observed in the R+ and the R- groups, but not in the NS group. Conversely, no significant difference for either score of immunoreactive cells were found between R+ and R- groups. The immunosuppressive regimen is suspected to play to play a major role in this depletion of bronchial mucosal T-cells. The acute lung rejection process does not appear to affect concurrently the immunohistochemical profile of immunoreactive cells in the bronchial mucosa.

Effectiveness and safety of bolus administration of alteplase in massive pulmonary embolism

Animal studies have demonstrated that thrombolysis with recombinant tissue-type plasminogen activator (rt-PA) is accelerated and that bleeding is reduced when rt-PA is infused over a short period. Previous clinical studies in patients with venous thromboembolism have shown that rt-PA is an effective thrombolytic agent when administered by continuous infusion over 2 to 24 hours. Clinical experience of bolus rt-PA administration in patients with massive acute pulmonary embolism (PE) is, however, limited. A prospective open study was conducted in which 54 patients with massive PE (Miller index > or = 20 of 34) received a 10-minute infusion of rt-PA at a dose of 1 mg/kg. Perfusion lung scanning was used to assess the change in pulmonary perfusion after drug administration. At 48 hours and 10 days, the mean absolute improvements in the perfusion defect were 11 and 31%, respectively. In addition, a significant clinical improvement occurred within 2 hours in 11 of the 15 shocked patients. Five patients died (9%) as a result of persistent shock (3 patients), neurologic damage (1 patient) or intracranial bleeding (1 patient). Major bleeding occurred in 8 patients (15%). Long-term follow-up information was available for 44 of the 49 discharged patients: 2 had died and 12 (27%) complained of persistent exertional dyspnea, 7 of whom had an associated heart or lung disease or chronic thromboembolism at admission. These results suggest that a bolus regimen of rt-PA could provide a convenient approach to thrombolytic therapy in patients with massive PE.(ABSTRACT TRUNCATED AT 250 WORDS)