Automated Affect Detection in Deep Brain Stimulation for Obsessive-Compulsive Disorder: A Pilot Study.

Automated measurement of affective behavior in psychopathology has been limited primarily to screening and diagnosis. While useful, clinicians more often are concerned with whether patients are improving in response to treatment. Are symptoms abating, is affect becoming more positive, are unanticipated side effects emerging? When treatment includes neural implants, need for objective, repeatable biometrics tied to neurophysiology becomes especially pressing. We used automated face analysis to assess treatment response to deep brain stimulation (DBS) in two patients with intractable obsessive-compulsive disorder (OCD). One was assessed intraoperatively following implantation and activation of the DBS device. The other was assessed three months post-implantation. Both were assessed during DBS on and o conditions. Positive and negative valence were quantified using a CNN trained on normative data of 160 non-OCD participants. Thus, a secondary goal was domain transfer of the classifiers. In both contexts, DBS-on resulted in marked positive affect. In response to DBS-off, affect flattened in both contexts and alternated with increased negative affect in the outpatient setting. Mean AUC for domain transfer was 0.87. These findings suggest that parametric variation of DBS is strongly related to affective behavior and may introduce vulnerability for negative affect in the event that DBS is discontinued.

Facing Imbalanced Data Recommendations for the Use of Performance Metrics.

Recognizing facial action units (AUs) is important for situation analysis and automated video annotation. Previous work has emphasized face tracking and registration and the choice of features classifiers. Relatively neglected is the effect of imbalanced data for action unit detection. While the machine learning community has become aware of the problem of skewed data for training classifiers, little attention has been paid to how skew may bias performance metrics. To address this question, we conducted experiments using both simulated classifiers and three major databases that differ in size, type of FACS coding, and degree of skew. We evaluated influence of skew on both threshold metrics (Accuracy, F-score, Cohen's kappa, and Krippendorf's alpha) and rank metrics (area under the receiver operating characteristic (ROC) curve and precision-recall curve). With exception of area under the ROC curve, all were attenuated by skewed distributions, in many cases, dramatically so. While ROC was unaffected by skew, precision-recall curves suggest that ROC may mask poor performance. Our findings suggest that skew is a critical factor in evaluating performance metrics. To avoid or minimize skew-biased estimates of performance, we recommend reporting skew-normalized scores along with the obtained ones.

Occupational transmission of Mycobacterium tuberculosis to health care workers in a university hospital in Lima, Peru.

From November 1996 through March 1997, presumptive active pulmonary tuberculosis (TB) was detected in 44 health care workers (HCWs) at a university hospital in Lima, Peru. To further assess the magnitude of the outbreak and determine risk factors for occupational Mycobacterium tuberculosis transmission, we identified HCWs in whom active pulmonary TB was diagnosed from January 1994 through January 1998, calculated rates by year and hospital work area, and conducted a tuberculin skin test (TST) survey. Thirty-six HCWs had confirmed active pulmonary TB. The rate of TB was significantly higher among the 171 HCWs employed in the laboratory than among HCWs employed in all other areas. In multivariate analysis, the only independent risk factor for HCW M. tuberculosis infection in HWCs employed in the laboratory was the use of common staff areas. Very high rates of active pulmonary TB were detected among HCWs at this hospital, and occupational acquisition in the laboratory was associated with HCW-to-HCW transmission.

Drug-resistant tuberculosis in the Dominican Republic: results of a nationwide survey.

SETTING: The Dominican Republic. OBJECTIVE: To assess the extent of drug-resistant tuberculosis (TB) following the guidelines of the World Health Organization (WHO)/International Union Against Tuberculosis and Lung Disease (IUATLD) new global surveillance project on drug resistance in TB. METHODS: Using a multi-step proportional weighted approach, a sample of 688 sequential cases of smear positive pulmonary TB diagnosed between April 1994 and April 1995 was studied in six of the country's eight health regions. Pre-treatment sputum samples were cultured on Loewenstein-Jensen medium and drug susceptibility tests were performed using the economic variant of the proportion method. RESULTS: Of 420 cases with drug susceptibility results, resistance to one or more drugs was observed in 43.8%; resistance was found in 52.1% of 117 TB cases with a history of previous antituberculosis treatment and in 40.6% of 303 new TB cases. In five of the six health regions surveyed, > or = 41% of strains were resistant to one or more drugs. Multidrug resistance (MDR) to isoniazid and rifampicin with or without resistance to other drugs was found in 43 (10.2%) of 420 cases, including 6.6% of new TB cases. In five of the six health regions > or = 8% of strains were classified as MDR. Independent predictors of MDR-TB included being in the age group 25 to 44 years (odds ratio [OR] = 4.2, 95% confidence interval [Cl] 1.5, 11.6; P = 0.005), being aged 45 years and over (OR = 4.5, 95% CI 1.4, 14.4; P = 0.009), and having a prior history of TB (OR = 3.7, 95% CI 1.9, 7.4; P = 0.0001). CONCLUSION: The proportion of Mycobacterium tuberculosis strains resistant to one or more anti-TB drugs in the Dominican Republic is among the highest observed world-wide. The severity of the problem urgently requires the full implementation of TB control strategies endorsed by the WHO and the IUATID, which include political commitment to a National TB Program, case detection utilizing sputum-smear microscopy, directly observed treatment, regular drug supply, and standardised recording and reporting systems. Also, the sale of TB drugs in the private market should be controlled.

Association between an early humoral response to Mycobacterium tuberculosis antigens and later development of tuberculosis in human immunodeficiency virus-infected individuals.

OBJECTIVE: To evaluate the association between an early humoral response to Mycobacterium tuberculosis antigens and the later development of tuberculosis (TB) disease in human immunodeficiency virus (HIV)-infected individuals. METHODS: Using an ELISA test, IgG antibodies against 4 M. tuberculosis antigens--purified protein derivative (PPD); 2,3 diacyl trehalose (DAT); a lipooligosaccharide (LOS) and a trehalose dicarboxylic acid bis N,N-dioctadecylamide (BDA.TDA)--were measured in sera from 25 HIV-infected tuberculous patients and 52 HIV-infected persons without TB. RESULTS: With the DAT and LOS antigens, a positive result in sera obtained in the 12 months preceding the onset of TB was significantly associated with later development of TB. Using the BDA.TDA antigen, the same association was observed in sera collected during the 6 months before the diagnosis of TB. No significant association was found with the PPD antigen. CONCLUSIONS: These results suggest that specific antibody markers may be useful to evaluate the risk of active TB in HIV-infected individuals, and a helpful indicator for preventive treatment.