Machine Learning and Health Science Research: Tutorial.

Machine learning (ML) has seen impressive growth in health science research due to its capacity for handling complex data to perform a range of tasks, including unsupervised learning, supervised learning, and reinforcement learning. To aid health science researchers in understanding the strengths and limitations of ML and to facilitate its integration into their studies, we present here a guideline for integrating ML into an analysis through a structured framework, covering steps from framing a research question to study design and analysis techniques for specialized data types.

A robust CRISPR-Cas9-based fluorescent reporter assay for the detection and quantification of DNA double-strand break repair.

DNA double-strand breaks (DSBs) are highly cytotoxic lesions that can lead to chromosome rearrangements, genomic instability and cell death. Consequently, cells have evolved multiple mechanisms to efficiently repair DSBs to preserve genomic integrity. We have developed a DSB repair assay system, designated CDDR (CRISPR-Cas9-based Dual-fluorescent DSB Repair), that enables the detection and quantification of DSB repair outcomes in mammalian cells with high precision. CDDR is based on the introduction and subsequent resolution of one or two DSB(s) in an intrachromosomal fluorescent reporter following the expression of Cas9 and sgRNAs targeting the reporter. CDDR can discriminate between high-fidelity (HF) and error-prone non-homologous end-joining (NHEJ), as well as between proximal and distal NHEJ repair. Furthermore, CDDR can detect homology-directed repair (HDR) with high sensitivity. Using CDDR, we found HF-NHEJ to be strictly dependent on DNA Ligase IV, XRCC4 and XLF, members of the canonical branch of NHEJ pathway (c-NHEJ). Loss of these genes also stimulated HDR, and promoted error-prone distal end-joining. Deletion of the DNA repair kinase ATM, on the other hand, stimulated HF-NHEJ and suppressed HDR. These findings demonstrate the utility of CDDR in characterizing the effect of repair factors and in elucidating the balance between competing DSB repair pathways.

Monitoring bladder compliance using end filling detrusor pressure: Clinical results and related factors.

OBJECTIVE: To assess the clinical significance of low compliance bladder (LCB) in women with lower urinary tract symptoms. MATERIALS AND METHODS: Medical records of 1490 women undergoing videourodynamic studies (VUSs) were reviewed. Comprehensive medical histories, physical examinations, bladder diaries, and results of multichannel VUS were analyzed. This study adopted an end filling detrusor pressure (EFP) greater than 20 cmH2O to define LCB. RESULTS: Among the study patients (n = 1490), 9.1% were diagnosed with LCB using a cutoff value of 17.5 cmH2O, which had a sensitivity and specificity of 89% and 92.7%, respectively. Results of multivariate analysis indicated that age (p = 0.005), maximum cystometric capacity (MCC; p = 0.002), detrusor overactivity (DO; p = 0.001), pelvic organ prolapse (POP; p = 0.018), recurrent urinary tract infection (p = 0.001), and radical abdominal hysterectomy (RAH; p < 0.001) as independent prognostic factors. Furthermore, our study results indicate that the MCC, urinary tract infection, and a history of RAH have a positive correlation with LCB, whereas, age, POP, and DO have a negative correlation with LCB. CONCLUSION: Our idea using EFP (≥17.5 cmH2O) for screening women with LCB is feasible for clinical use.

Postvoid residual urine in women with stress incontinence.

AIMS: To confirm the idea that women with stress incontinence can have elevated postvoid residual urine (PVR) and to examine the correlation between PVR obtained with catheterization versus that with BladderScan (BS). MATERIALS AND METHODS: This is a prospective study involving 902 women referred to our urogynecology clinics because of symptoms of lower urinary tract dysfunction. Women were selected if they met all of the following conditions: (1) A main complaint of stress urinary incontinence; (2) A diagnosis of urodynamic stress incontinence; and (3) No previous pelvic surgery, advanced pelvic prolapse or neurological deficit. One hundred and seven women met all criteria and formed the basis for this study. All women in the study group underwent three-part urodynamic testing including uroflowmetry, filling (provocative) and voiding cystometry. After uroflowmetry they were scanned by a BS, and then catheterized for PVR volume before the procedure of cystometry. RESULTS: The mean PVR volume was 62.8 ml by BS and 38.5 ml by catheterization. 35.5% women had PVR urine higher than 50 ml and 15.9% had PVR urine greater than 100 ml. The PVR volume obtained by BS correlated significantly with catheterized volume (r = 0.625, P = 0.001) and offered a sensitivity of 64.7% and a specificity of 94.3% in detecting PVR greater than 100. The mean maximum flow rate was 22.1 ml/sec and mean detrusor contraction pressure during voiding was 21 cm H(2)O. CONCLUSIONS: Women in our study had low maximum flow rate (22.1 ml/sec), elevated PVR (38.5 ml) and high detrusor contraction pressure during voiding (21 cm H(2)O) indicating that women with stress incontinence have some degree of voiding dysfunction. The bladder behavior in women with stress incontinence may be more complex than we had previously considered and special care should be taken if a woman with elevated PVR is scheduled for anti-incontinence surgery. BS appears to be reasonably sensitive and specific for the detection of elevated PVR and is reliable in clinical use.