Evaluation of the Statistical Detection of Change Algorithm for Screening Patients with MS with New Lesion Activity on Longitudinal Brain MRI

BACKGROUND AND PURPOSE

Identification of new MS lesions on longitudinal MR imaging by human readers is time-consuming and prone to error. Our objective was to evaluate the improvement in the performance of subject-level detection by readers when assisted by the automated statistical detection of change algorithm.

MATERIALS AND METHODS

A total of 200 patients with MS with a mean interscan interval of 13.2 (SD, 2.4) months were included. Statistical detection of change was applied to the baseline and follow-up FLAIR images to detect potential new lesions for confirmation by readers (Reader + statistical detection of change method). This method was compared with readers operating in the clinical workflow (Reader method) for a subject-level detection of new lesions.

RESULTS

Reader + statistical detection of change found 30 subjects (15.0%) with at least 1 new lesion, while Reader detected 16 subjects (8.0%). As a subject-level screening tool, statistical detection of change achieved a perfect sensitivity of 1.00 (95% CI, 0.88-1.00) and a moderate specificity of 0.67 (95% CI, 0.59-0.74). The agreement on a subject level was 0.91 (95% CI, 0.87-0.95) between Reader + statistical detection of change and Reader, and 0.72 (95% CI, 0.66-0.78) between Reader + statistical detection of change and statistical detection of change.

CONCLUSIONS

The statistical detection of change algorithm can serve as a time-saving screening tool to assist human readers in verifying 3D FLAIR images of patients with MS with suspected new lesions. Our promising results warrant further evaluation of statistical detection of change in prospective multireader clinical studies.

Outcomes of intravenous tissue plasminogen activator for acute ischaemic stroke in HIV-infected adults

BACKGROUND AND PURPOSE

To our knowledge there are no studies reporting the use and short-term outcomes of intravenous tissue plasminogen activator (IV-TPA) for the treatment of acute ischaemic stroke (AIS) in people living with HIV.

METHODS

The US Nationwide Inpatient Sample (NIS) (2006-2010) was searched for HIV-infected AIS patients treated with IV-TPA.

RESULTS

In the NIS, 2.2% (62/2877) of HIV-infected AIS cases were thrombolyzed with IV-TPA (median age 52 years, range 27-78, 32% female, 22% Caucasian) vs. 2.1% (19 335/937 896) of HIV-uninfected cases (median age 72 years, range 17-102 years, 50% female, 74% Caucasian; P = 0.77). There were more deaths in HIV-infected versus uninfected patients with stroke (220/2877, 7.6% vs. 49 089/937 547, 5.2%, P < 0.001) but no difference in the proportion of deaths amongst IV-TPA-treated patients. The age- and sex-adjusted odds ratio for death following IV-TPA administration in HIV-infected versus uninfected patients was 2.26 (95% CI 1.12, 4.58), but the interaction on mortality between HIV and IV-TPA use was not statistically significant, indicating no difference in risk of in-hospital death by HIV serostatus with IV-TPA use. A higher number of HIV-infected patients remained in hospital versus died or were discharged at both 10 and 30 days (P < 0.01 at 10 and 30 days). No difference in the proportion of intracerebral hemorrhage in the two groups was found (P = 0.362).

CONCLUSIONS

The in-hospital mortality is higher amongst HIV-infected AIS patients than HIV-uninfected patients. However, the risk of death amongst HIV-infected patients treated with IV-TPA is similar to HIV-uninfected groups.

Automatic lesion incidence estimation and detection in multiple sclerosis using multisequence longitudinal MRI

BACKGROUND AND PURPOSE

Detecting incidence and enlargement of lesions is essential in monitoring the progression of MS. In clinical trials, lesion load is observed by manually segmenting and comparing serial MR images, which is time consuming, costly, and prone to inter- and intraobserver variability. Subtracting images from consecutive time points nulls stable lesions, leaving only new lesion activity. We propose SuBLIME, an automated method for segmenting incident lesion voxels.

MATERIALS AND METHODS

We used logistic regression models incorporating multiple MR imaging sequences and subtraction images from consecutive longitudinal studies to estimate voxel-level probabilities of lesion incidence. We used T1-weighted, T2-weighted, FLAIR, and PD volumes from a total of 110 MR imaging studies from 10 subjects.

RESULTS

To assess the performance of the model, we assigned 5 subjects to a training set and the remaining 5 to a validation set. With SuBLIME, lesion incidence is detected and delineated in the validation set with an AUC of 99% (95% CI [97%, 100%]) at the voxel level.

CONCLUSIONS

This fully automated and computationally fast method allows sensitive and specific detection of lesion incidence that can be applied to large collections of images. Using the explicit form of the statistical model, SuBLIME can easily be adapted to cases when more or fewer imaging sequences are available.

A comparison of indices of respiratory failure in ventilated preterm infants

AIM

To compare indices of respiratory failure in terms of their ability to predict adverse respiratory outcomes in preterm infants. The indices evaluated were: (a) the alveolar-arterial oxygen tension difference (A-aDO(2)); (b) the ratio of arterial to alveolar oxygen tension (a/A ratio); (c) the oxygenation index (OI); (d) the fractional inspired oxygen concentration (FIO(2)).

METHODS

Details of respiratory support and arterial blood gas data in the first 24 hours of life were collected in ventilated infants below 34 weeks gestation. The worst single value of a particular index in the first 24 hours was chosen to quantify the severity of respiratory failure in each infant. Receiver operating characteristic curves were constructed and areas under the curve (AUC) calculated to compare the performance of the indices in predicting death from respiratory failure and/or the development of chronic lung disease (CLD).

RESULTS

A total of 155 preterm infants were studied, of whom 35 (23%) died primarily from respiratory failure and 53 of the 120 survivors (44%) developed CLD. The overall performance of the four indices in predicting death from respiratory failure ranged from 0. 77 (AUC for maximum FIO(2)) to 0.88 (AUC for minimum a/A ratio). The corresponding AUCs for gestational age and birth weight were 0.75 and 0.76 respectively. In contrast, demographic variables tended to perform better than indices of respiratory failure in predicting CLD/death.

CONCLUSIONS

There was no evidence of a significant difference between the performance of the a/A ratio, A-aDO(2), and OI in predicting adverse respiratory outcomes. Use of the OI is recommended because of its ease of calculation.

Mothers with mental illness: II. Family relationships and the context of parenting

OBJECTIVE

This study explores the experiences of mothers with mental illness regarding their family relationships.

METHODS

Six focus groups of mothers with mental illness and five focus groups of case managers met to discuss problems facing mothers with mental illness and to recommend solutions. Focus groups were audiotaped, and transcripts were coded and analyzed qualitatively to describe ways in which husbands and partners, grandparents, and other family members contribute to the context of parenting for mothers with mental illness.

RESULTS

Mothers with mental illness and case managers described a range of relationships and attitudes of family members and provided examples of the ways in which family members contribute both positively and negatively to parenting.

CONCLUSIONS

Although family members may seem to be natural supports for mothers with mental illness, their involvement in the context of parenting may not be entirely positive. The contributions of family members to the context of parenting for mothers with mental illness must be considered by treatment providers if unintentional negative outcomes are to be avoided.

Mothers with mental illness: I. The competing demands of parenting and living with mental illness

OBJECTIVE

The purpose of this study is to understand the parenting experiences of women with mental illness from the perspectives of mothers and case managers employed by the state department of mental health.

METHODS

Six focus groups of mothers and five focus groups of case managers met to discuss the problems facing mothers with mental illness and to recommend solutions. Focus-group transcripts were coded and items grouped by themes in qualitative analyses to explore the conflicts mothers face in meeting the dual challenges of parenting and living with mental illness.

RESULTS

Mothers and case managers identified sources of conflict in four thematic categories: the stigma of mental illness, day-to-day parenting, managing mental illness, and custody of and contact with children.

CONCLUSIONS

Many of the issues of mothers with mental illness are generic to all parents; others are specific to the situation of living with mental illness. Mothers with mental illness must play a role in developing standards for clinical care and the research agenda in this area.