Machine learning for microbial identification and antimicrobial susceptibility testing on MALDI-TOF mass spectra: a systematic review.

BACKGROUND: The matrix assisted laser desorption/ionization and time-of-flight mass spectrometry (MALDI-TOF MS) technology has revolutionized the field of microbiology by facilitating precise and rapid species identification. Recently, machine learning techniques have been leveraged to maximally exploit the information contained in MALDI-TOF MS, with the ultimate goal to refine species identification and streamline antimicrobial resistance determination. OBJECTIVES: The aim was to systematically review and evaluate studies employing machine learning for the analysis of MALDI-TOF mass spectra. DATA SOURCES: Using PubMed/Medline, Scopus and Web of Science, we searched the existing literature for machine learning-supported applications of MALDI-TOF mass spectra for microbial species and antimicrobial susceptibility identification. STUDY ELIGIBILITY CRITERIA: Original research studies using machine learning to exploit MALDI-TOF mass spectra for microbial specie and antimicrobial susceptibility identification were included. Studies focusing on single proteins and peptides, case studies and review articles were excluded. METHODS: A systematic review according to the PRISMA guidelines was performed and a quality assessment of the machine learning models conducted. RESULTS: From the 36 studies that met our inclusion criteria, 27 employed machine learning for species identification and nine for antimicrobial susceptibility testing. Support Vector Machines, Genetic Algorithms, Artificial Neural Networks and Quick Classifiers were the most frequently used machine learning algorithms. The quality of the studies ranged between poor and very good. The majority of the studies reported how to interpret the predictors (88.89%) and suggested possible clinical applications of the developed algorithm (100%), but only four studies (11.11%) validated machine learning algorithms on external datasets. CONCLUSIONS: A growing number of studies utilize machine learning to optimize the analysis of MALDI-TOF mass spectra. This review, however, demonstrates that there are certain shortcomings of current machine learning-supported approaches that have to be addressed to make them widely available and incorporated them in the clinical routine.

Normative data of contact heat evoked potentials from the lower extremities.

Contact heat evoked potentials (CHEPs) have become an acknowledged research tool in the assessment of the integrity of the nociceptive system and gained importance in the diagnostic work-up of patients with suspected small fiber neuropathy. For the latter, normative values for CHEP amplitude and latency are indispensable for a clinically meaningful interpretation of the results gathered in patients. To this end, CHEPs were recorded in 100 healthy subjects over a wide age range (20-80 years) and from three different dermatomes of the lower extremities (L2, L5, and S2). A normal baseline (35-52 °C) and increased baseline stimulation (42-52 °C) were applied. Statistical analysis revealed significant effects of stimulation site, stimulation intensity, and sex on CHEP parameters (N2 latency, N2P2 amplitude, and NRS). Significant positive correlations of body height with N2 latency, and pain ratings with N2P2 amplitudes were observed. This is the first time that normative values have been obtained from multiple dermatomes of the lower extremities. The present dataset will facilitate the clinical application of CHEPs in the neurophysiological diagnosis of small fiber neuropathy and by discerning pathological findings help establish a proximal-distal gradient of nerve degeneration in polyneuropathies.

Recent advances in objectifying pain using neuroimaging techniques.

The pursuit of a physiological indicator of noxious stimulation is desirable as it has the potential to provide mechanistic information regarding acute pain and may ultimately improve pain management strategies. Currently, there are no specific neurophysiological markers of pain to evaluate treatments. Recent attempts to identify neural correlates of pain have focused on different neuroimaging modalities. The purpose of this review is to discuss common neuroimaging techniques and findings thus far.

Contact heat evoked potentials: Reliable acquisition from lower extremities.

OBJECTIVE: To investigate test-retest reliability of contact heat evoked potentials (CHEPs) from lower extremities using two different stimulation protocols, i.e., normal and increased baseline temperature. METHODS: A total of 32 able-bodied subjects were included and a subset (N = 22) was retested. CHEPs were recorded from three different dermatomes of the lower extremity (i.e., L2, L5, and S2). Test-retest reliability of CHEPs acquisition after simulation in various lower limb dermatomes using different stimulation protocols was analyzed. RESULTS: The study revealed an improved acquisition of CHEPS employing the increased baseline protocol, particularly when stimulating more distal sites, i.e., dermatome L5 and S2. Based on repeatability coefficients, CHEP latency (N2 potential) emerged as the most robust CHEP parameter. Although CHEP amplitudes (N2P2 complex) and pain ratings were decreased in the retest, amplitudes still showed fair to excellent intraclass correlation coefficients using normal baseline or increased baseline temperature, respectively. CONCLUSIONS: This is the first study to demonstrate that CHEPs acquisition from the lower extremities is improved by increasing the baseline temperature of the thermode. SIGNIFICANCE: This study highlights the usability of CHEPs as a viable diagnostic method to study small fiber integrity.

Discrepancy between perceived pain and cortical processing: A voxel-based morphometry and contact heat evoked potential study.

OBJECTIVES: The purpose of this study was to determine if local gray and white matter volume variations between subjects could account for variability in responses to CHEP stimulation. METHODS: Structural magnetic resonance imaging was used to perform voxel-based morphometry (VBM) of gray and white matter in 30 neurologically healthy subjects. Contact heat stimulation was performed on the dorsum of the right hand at the base of the thumb. Evoked potentials were acquired from a vertex-recording electrode referenced to linked ears. RESULTS: Controlling for age, total intracranial volume, and skull/scalp thickness, CHEP amplitude and pain rating were not significantly correlated between subjects. A VBM region of interest approach demonstrated a significant interaction between pain rating and N2 amplitude in the right insular cortex (p<0.05, family-wise error corrected, FWE). In white matter, a significant interaction was localized in the right inferior frontal occipital fasciculus (IFOF, p<0.05 FWE). CONCLUSIONS: Accounting for gray matter volume in the right insular cortex, resulted in a significant relationship between CHEP amplitude and pain rating. SIGNIFICANCE: This finding suggests that the discrepancy between pain ratings and the amplitude of evoked potentials is not solely related to measurement artifact, but rather attributable, in part, to anatomical differences between subjects.

Relationship between chronic pain and brain reorganization after deafferentation: A systematic review of functional MRI findings.

BACKGROUND: Mechanisms underlying the development of phantom limb pain and neuropathic pain after limb amputation and spinal cord injury, respectively, are poorly understood. The goal of this systematic review was to assess the robustness of evidence in support of "maladaptive plasticity" emerging from applications of advanced functional magnetic resonance imaging (MRI). METHODS: Using MeSH heading search terms in PubMed and SCOPUS, a systematic review was performed querying published manuscripts. RESULTS: From 146 candidate publications, 10 were identified as meeting the inclusion criteria. Results from fMRI investigations provided some level of support for maladaptive cortical plasticity, including longitudinal studies that demonstrated a change in functional organization related to decreases in pain. However, a number of studies have reported no relationship between reorganization, pain and deafferentation, and emerging evidence has also suggested the opposite - that is, chronic pain is associated with preserved cortical function. CONCLUSION: Based solely on advanced functional neuroimaging results, there is only limited evidence for a relationship between chronic pain intensity and reorganization after deafferentation. The review demonstrates the need for additional neuroimaging studies to clarify the relationship between chronic pain and reorganization.