Discriminating between bipolar and major depressive disorder using a machine learning approach and resting-state EEG data.

OBJECTIVE: Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge as effective treatment is quite different for each condition. In this study electroencephalography (EEG) was explored as an objective biomarker for distinguishing MDD from BD using an efficient machine learning algorithm (MLA) trained by a relatively large and balanced dataset. METHODS: A 3 step MLA was applied: (1) a multi-step preprocessing method was used to improve the quality of the EEG signal, (2) symbolic transfer entropy (STE), an effective connectivity measure, was applied to the resultant EEG and (3) the MLA used the extracted STE features to distinguish MDD (N = 71) from BD (N = 71) subjects. RESULTS: 14 connectivity features were selected by the proposed algorithm. Most of the selected features were related to the frontal, parietal, and temporal lobe electrodes. The major involved regions were the Broca region in the frontal lobe and the somatosensory association cortex in the parietal lobe. These regions are near electrodes FC5 and CPz and are involved in processing language and sensory information, respectively. The resulting classifier delivered an evaluation accuracy of 88.5% and a test accuracy of 89.3%, using 80% of the data for training and evaluation and the remaining 20% for testing, respectively. CONCLUSIONS: The high evaluation and test accuracies of our algorithm, derived from a large balanced training sample suggests that this method may hold significant promise as a clinical tool. SIGNIFICANCE: The proposed MLA may provide an inexpensive and readily available tool that clinicians may use to enhance diagnostic accuracy and shorten time to effective treatment.

A Machine Learning Algorithm to Discriminating Between Bipolar and Major Depressive Disorders Based on Resting EEG Data.

Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge due to the lack of known biomarkers. Conventional methods of diagnosis rest exclusively on symptomatic presentation, and personal and family history. As a result, BD-depressed episode (BD-DE) is often misdiagnosed as MDD, and inappropriate therapy is given. Electroencephalography (EEG) has been widely studied as a potential source of biomarkers to differentiate these disorders. Previous attempts using machine learning (ML) methods have delivered insufficient sensitivity and specificity for clinical use, likely as a consequence of the small training set size, and inadequate ML methodology. We hope to overcome these limitations by employing a training dataset of resting-state EEG from 71 MDD and 71 BD patients. We introduce a robust 3 steps ML technique: 1) a multi-step preprocessing method is used to improve the quality of the EEG signal 2) symbolic transfer entropy (STE), which is an effective connectivity measure, is applied to the resultant EEG signals 3) the ML algorithm uses the extracted STE features to distinguish MDD from BD patients. Clinical Relevance--- The accuracy of our algorithm, derived from a large sample of patients, suggests that this method may hold significant promise as a clinical tool. The proposed method delivered total accuracy, sensitivity, and specificity of 84.9%, 83.4%, and 87.1%, respectively.

Survival outcome of cervical cancer patients treated by image-guided brachytherapy: a 'real world' single center experience in Thailand from 2008 to 2018.

The objective of our study was to evaluate the survival outcome of cervical cancer patients treated using image-guided brachytherapy (IGBT). From 2008 to 2018, 341 patients with cervical cancer were treated by radical radiotherapy. IGBT (by computed tomography [CT] or transabdominal ultrasound [TAUS]) was used to treat all of these patients. The characteristic data and patient status after treatment were recorded. All data were evaluated for survival outcome analysis. From a total of 341 patients, 295 patients were analyzed and 46 patients were excluded due to data missing in the survival outcomes. At the median follow-up time of 48 months (IQR 30-80 months), The 4-year local control, progression-free survival and overall survival rates were 89.5%, 74.9% and 69.1%, respectively. For overall survival, the size (> 5 cm), pathology (non-SCCA), stage (stage III-IV by FIGO 2009), lymph node (LN) (presented) and overall treatment time (OTT) (> 56 days) showed statistical significance in univariate analysis while non-SCCA pathology, advanced stage, presented LN and longer OTT showed statistical significance in multivariate analysis. In conclusion, our analysis reports a 4-year overall survival rate of 69.1%. Non-SCCA pathology, advanced stage disease, LN presence and longer OTT showed worse prognostic factors in multivariate analysis.

High-frequency burst vagal nerve simulation therapy in a natural primate model of genetic generalized epilepsy.

PURPOSE: Since the approval of Vagal Nerve Stimulation (VNS) Therapy for medically refractory focal epilepsies in 1997, it has been also reported to be effective for a wide range of generalized seizures types and epilepsy syndromes. Instead of conventional VNS Therapy delivered at 20-30Hz signal frequencies, this study evaluates efficacy and tolerability of high-frequency burst VNS in a natural animal model for genetic generalized epilepsy (GGE), the epileptic baboon. METHODS: Two female baboons (B1 P.h. Hamadryas and B2 P.h. Anubis x Cynocephalus) were selected because of frequently witnessed generalized tonic-clonic seizures (GTCS) for VNS implantation. High-frequency burst VNS Therapy was initiated after a 4-5 week baseline; different VNS settings (0.25, 2 or 2.5mA, 300Hz, 4 vs 7 pulses, 0.5-2.5s interburst interval, and intermittent stimulation for 1-2 vs for 24h per day) were tested over the subsequent 19 weeks, which included a 4-6 week wash-out period. GTCS frequencies were quantified for each setting, while seizure duration and postictal recovery times were compared to baseline. Scalp EEG studies were performed at almost every setting, including intermittent light stimulation (ILS) to evaluate photosensitivity. Pre-ILS ictal and interictal discharge rates, as well as ILS responses were compared between trials. The Novel Object test was used to assess potential treatment effects on behavior. RESULTS: High-frequency burst VNS Therapy reduced GTCS frequencies at all treatment settings in both baboons, except when output currents were reduced (0.25mA) or intermittent stimulation was restricted (to 1-2h/day). Seizure duration and postictal recovery times were unchanged. Scalp EEG studies did not demonstrate treatment-related decrease of ictal or interictal epileptic discharges or photosensitivity, but continuous treatment for 120-180s during ILS appeared to reduce photoparoxysmal responses. High-frequency burst VNS Therapy was well-tolerated by both baboons, without cardiac or behavioral changes. Repetitive muscle contractions involving the neck and left shoulder girdle were observed intermittently, most commonly at 0.5 interburst intervals, but these were transient, resolving with a few cycles of stimulation and not noted in wakefulness. CONCLUSIONS: This preclinical pilot study demonstrates efficacy and tolerability of high-frequency burst VNS Therapy in the baboon model of GGE. The muscle contractions may be due to aberrant propagation of the stimulus along the vagal nerve or to the ansa cervicalis, but can be reduced by minimal adjustment of current output or stimulus duration.

[Late onset visual aura associated with a carotid artery stenosis reversible after endarterectomy]. / Migraine avec aura à début tardif résolutive après endartérectomie sur sténose carotide serrée.

BACKGROUND: Visual aura is the most common type of aura. Onset of visual aura occurring for the first time in an elderly person is atypical, even though late-life migraine accompaniments are well-recognized. OBSERVATION: We report the case of a 72-year-old woman with a past history of migraine without aura since adolescence (one attack per month), who experienced for the first time of her life multiple visual aura over a period of more than one year. These auras were always left homonymous hemianopia which developed gradually over 5 mins followed by severe migraine headache without strict lateralization. The patient was known to have a moderate right carotid artery stenosis. In October 2007, she suffered a minor ischemic stroke in the right carotid artery territory due to progression to severe stenosis of the right internal carotid artery. After carotid endarterectomy, she had no further migraine attacks with aura. DISCUSSION: We discuss the mechanism of aura in our case with probable microemboli due to the severe right internal carotid artery stenosis responsible for repeated cortical spreading depression. CONCLUSION: Clinicians should bear in mind to systematically explore all patients with late onset aura including patients who have previously suffered from migraine without aura and also to explore patients who have a dramatic increase of aura even if they are already migrainers with aura.