Attempts to replicate the claim of James' at NSLS beam line X12B
Attempts to replicate the claim of James' at NSLS beam line X12B
复制标题
尝试复制 James 在 NSLS 光束线 X12B 上的主张
DOI：
10.1080/08940889908261034
复制
发表时间：
1999
影响因子：
--
通讯作者：
M. Capel
中科院分区：
文献类型：
--
作者：
K. Schroer;N. Abumrad;D. Derisi;Karen Kastrow;Erna Busch;N. Volkow;M. Capel
我们试图验证James等人的主张，即毛发图案的小角X射线衍射的特定变化与乳腺癌的病理学发现明确相关[I]。从正常和阳性对照（先前活检和诊断的患者）和进行组织活检以测定疑似乳腺癌的受试者获得阴毛样品。采用盲法方案进行采样。45名受试者提供了头发样本，但由于头发长度不合适，只有23名患者的样本适合分析。在对衍射图进行分类之前，实验者无法获得关于受试者疾病状态的知识。揭盲后，发现在分析的23份患者样本中，8份来自乳腺癌患者（诊断为原位或浸润性导管癌的乳腺活检样本的阳性病理学分析），15份来自对照组（乳腺活检样本的阴性病理学分析）。在NSLS光束线X12 B处使用300 × 300微米光束，波长= 1. 38埃，照相机距离= 491 mm（He吹扫飞行路径，样品气隙-2 cm）。用ADSC Quantum-4 CCD区域检测器（每个样品3 × 300秒去锯齿曝光）测量衍射图案。如下制备毛发样品：使用双面kapton胶带将每个受试者的4至7根毛发在lexan分裂垫圈上仔细对齐。然后将分裂垫圈安装在晶体测角器头上并对齐，使得毛发阵列的平行度相对于光束轴的偏差小于0.02度（毛发阵列沿着光束轴的总宽度< 600微米，阵列平坦度与毛发厚度相称，考虑到其扭结倾向）。在将样品阵列固定到垫圈的胶带表面上时，我们小心地避免在构成阵列的单个毛发中引起应变或扭曲。对于每个头发样品，将3个响应校正的300秒CCD图像相加，并沿衍射图像平面的赤道轴和赤道轴沿着进行扇区积分（扇区宽度30度，环形箱宽度1像素宽度）。根据James [1]，针对背景校正扇形积分，并测定是否存在对应于45埃d间距的近似各向同性衍射环。图1显示了“代表性”阴性诊断受试者的赤道（上）和赤道（下）扇区积分图。图2显示了阳性诊断受试者的相同迹线。q= 0.138 k1处的垂直线定位推定诊断峰的位置。扇区积分的分类是每-
We have attempted to validate the claim of James et al. that specific changes in the small angle X-ray diffraction of pattern of hair are unambiguously correlated with the pathological finding of breast cancer [I]. Pubic hair samples were obtained from normal and positive controls (previously biopsied and diagnosed patients) and subjects undergoing tissue biopsy for assay of suspected breast cancer. Sampling was conducted using a blind protocol. Forty-five subjects contributed hair samples but because of inappropriate hair length only the samples from 23 of the patients were suitable for analysis. No knowledge concerning the disease state of the subjects was available to the experimenters prior to categorization of the diffraction patterns. After breaking of the blinding code, it was found that, of the 23 patient samples analyzed, 8 were from patients with breast cancer (positive pathological analysis of breast biopsy samples with a diagnosis of in situ or invasive ductal carcinoma) and 15 were from controls (negative pathological analysis of breast biopsy samples). Low-angle diffraction patterns were measured at NSLS beam line X12B using a 300x300 micron beam, wavelength= l. 38 angstroms, camera distance= 491 mm (He-purged flight path, sample air gap-2 cm). Diffraction patterns were measured with an ADSC Quantum-4 CCD area detector (3x300 sec de-zingered exposures per sample). Hair samples were prepared as follows: four to seven hairs per subject were carefully aligned on a lexan split washer using doublesided kapton tape. The split washer was then mounted on a crystallographic goniometer head and aligned so that deviation from perpendicularity of the hair array was less than 0.02 degrees relative to the beam axis (total width of hair array along beam axis< 600 micron, array flatness commensurate with hair thickness, given its tendency for kinking). In securing the sample array to the taped surface of the washer, we were careful not to induce strain or twist in the individual hairs comprising the array. Prior to mounting and exposure, all samples were contained in a sealed jar (-24 hours) to assure constant humidity for all samples.For each hair sample, the 3 response-corrected 300-second CCD images were summed and sectorially integrated along the equatorial and meridional axes of the diffraction image plane (sector width 30 degrees, annular bin width 1 pixel width). Sector integrals were corrected for background and assayed for the presence or absence of an approximately isotropic diffraction ring corresponding to a d-spacing of 45 angstroms, according to James [l]. Figure 1 shows the plots of the equatorial (upper) and meridional (lower) sector integrals for a “representative” negatively diagnosed subject. Figure 2 shows the same traces for a positively diagnosed subject. The vertical line at q= 0.138 k1 locates the position of the putative diagnostic peak. Categorization of the sector integrals was per-