Deep learning provides high accuracy in automated chondrocyte viability assessment in articular cartilage using nonlinear optical microscopy.

Chondrocyte viability is a crucial factor in evaluating cartilage health. Most cell viability assays rely on dyes and are not applicable for in vivo or longitudinal studies. We previously demonstrated that two-photon excited autofluorescence and second harmonic generation microscopy provided high-resolution images of cells and collagen structure; those images allowed us to distinguish live from dead chondrocytes by visual assessment or by the normalized autofluorescence ratio. However, both methods require human involvement and have low throughputs. Methods for automated cell-based image processing can improve throughput. Conventional image processing algorithms do not perform well on autofluorescence images acquired by nonlinear microscopes due to low image contrast. In this study, we compared conventional, machine learning, and deep learning methods in chondrocyte segmentation and classification. We demonstrated that deep learning significantly improved the outcome of the chondrocyte segmentation and classification. With appropriate training, the deep learning method can achieve 90% accuracy in chondrocyte viability measurement. The significance of this work is that automated imaging analysis is possible and should not become a major hurdle for the use of nonlinear optical imaging methods in biological or clinical studies.

Effects of intermittent IL-2 alone or with peri-cycle antiretroviral therapy in early HIV infection: the STALWART study.

BACKGROUND: The Study of Aldesleukin with and without antiretroviral therapy (STALWART) evaluated whether intermittent interleukin-2 (IL-2) alone or with antiretroviral therapy (ART) around IL-2 cycles increased CD4(+) counts compared to no therapy. METHODOLOGY: Participants not on continuous ART with > or = 300 CD4(+) cells/mm(3) were randomized to: no treatment; IL-2 for 5 consecutive days every 8 weeks for 3 cycles; or the same IL-2 regimen with 10 days of ART administered around each IL-2 cycle. CD4(+) counts, HIV RNA, and HIV progression events were collected monthly. PRINCIPAL FINDINGS: A total of 267 participants were randomized. At week 32, the mean CD4(+) count was 134 cells greater in the IL-2 alone group (p<0.001), and 133 cells greater in the IL-2 plus ART group (p<0.001) compared to the no therapy group. Twelve participants in the IL-2 groups compared to 1 participant in the group assigned to no therapy experienced an opportunistic event or died (HR 5.84, CI: 0.59 to 43.57; p = 0.009). CONCLUSIONS: IL-2 alone or with peri-cycle HAART increases CD4(+) counts but was associated with a greater number of opportunistic events or deaths compared to no therapy. These results call into question the immunoprotective significance of IL-2-induced CD4(+) cells. TRIAL REGISTRATION: ClinicalTrials.gov NCT00110812.

HIV replication capacity is an independent predictor of disease progression in persons with untreated chronic HIV infection.

OBJECTIVE: To assess the effect of pol replication capacity (RC) on the hazard ratio of progression to a composite endpoint of time to progression to <350 CD4+ cells per microliter, initiation of therapy, or death. METHODS: pol RC assays were performed after study closure in baseline samples obtained from 316 enrollees in a prospectively monitored cohort of treatment-naive adults with >or=450 CD4+ cells per microliter and >or=1000 HIV-1 RNA copies per milliliter. RESULTS: The median RC was 79%. Patients with a lower RC had a lower median viral load (4.0 vs 4.2 Log HIV-1 RNA copies/mL, P = 0.026) and a lower rate of protease inhibitor resistance 2% vs 8%, P = 0.03). Otherwise, baseline demographic and laboratory characteristics were similar. The hazard ratio of progression to the composite endpoint was 0.73 (P = 0.041) for persons with lower RC, 2.07 per 1.0 log10 higher viral load (P < 0.001), and 0.86 per 50 cells per microliter higher CD4+ cell count (P < 0.001). The effect of lower RC was also significant in a separate analysis of time to initiation of therapy (P = 0.04). CONCLUSIONS: These results show that untreated patients with lower vs higher RC had a slower rate of progression as assessed by a composite outcome of time to CD4+ count