新生兒局灶性白質(zhì)損傷的MRI研究現(xiàn)狀 及其核心問題的思考

楊 健，李賢軍

(西安交通大學(xué)第一附屬醫(yī)院醫(yī)學(xué)影像科，陜西西安 710061)

世界衛(wèi)生組織公布的數(shù)據(jù)顯示，我國早產(chǎn)兒數(shù)量居高不下。隨著兩孩政策的實(shí)施，高齡產(chǎn)婦數(shù)量增加，早產(chǎn)兒基數(shù)將進(jìn)一步增大。早產(chǎn)兒腦部發(fā)育不成熟，極易受到各類致病因素的影響而出現(xiàn)腦損傷，其中局灶性白質(zhì)損傷(punctate white matter lesions, PWML)在臨床實(shí)踐中最為常見。多項(xiàng)報(bào)道指出，PWML在新生兒中的發(fā)生率高達(dá)20%[1-3]。PWML可導(dǎo)致腦癱、弱視等不良發(fā)育結(jié)局[1]，對(duì)其進(jìn)行早期診斷、精確診斷與預(yù)后評(píng)估對(duì)于提升我國人口素質(zhì)至關(guān)重要。面對(duì)我國進(jìn)入老齡化社會(huì)的現(xiàn)狀，研發(fā)和應(yīng)用適宜于新生兒影像學(xué)檢查的新技術(shù)，對(duì)優(yōu)生優(yōu)育和實(shí)現(xiàn)強(qiáng)國發(fā)展戰(zhàn)略具有高度的迫切性和重大意義。

目前對(duì)PWML診斷的首選方法是磁共振成像(magnetic resonance imaging, MRI)技術(shù)[1-3]，因其具有無創(chuàng)、無輻射以及較高的空間分辨能力等優(yōu)點(diǎn)，適用于新生兒的顱腦檢查。在T1加權(quán)成像(T1 weighted imaging, T1WI)與T2加權(quán)成像(T2 weighted imaging, T2WI)等常規(guī)MRI圖像上，PWML表現(xiàn)為局灶性的點(diǎn)狀、線狀或簇狀異常信號(hào)(T1WI高信號(hào)，T2WI低信號(hào)或等信號(hào))[2-3]。除了以上常規(guī)MRI圖像上肉眼可見的顯性病變以外，PWML往往伴隨著微結(jié)構(gòu)隱性病變。隱性病變可能是顯性病變的繼發(fā)變化或伴隨發(fā)生，并且具有不易探測(cè)和分布廣泛的特點(diǎn)，而發(fā)育結(jié)局恰恰與腦損傷的分布范圍密切相關(guān)[4-5]。

1 國內(nèi)外研究現(xiàn)狀

1.1 PWML的損傷類型及其空間分布 腦白質(zhì)損傷類型及其空間分布是對(duì)PWML準(zhǔn)確診斷的基礎(chǔ)[4-6]。其損傷類型主要分為出血型和非出血型，且以非出血型為主，其磁敏感加權(quán)成像(susceptibility weighted imaging, SWI)表現(xiàn)為高信號(hào)，而與出血型的低信號(hào)易于鑒別(圖1)。PWML的分布很有規(guī)律性，無論在早產(chǎn)兒還是足月兒，深部白質(zhì)的損傷位置均圍繞在側(cè)腦室周圍，以側(cè)腦室后角旁、半卵圓中心區(qū)白質(zhì)最為常見(占63.64%)，可能是因?yàn)樗枨拾l(fā)育遵循自背側(cè)向腹側(cè)、自中心向四周的規(guī)律，該區(qū)髓鞘化更活躍故易受損傷；與其他研究認(rèn)為該信號(hào)改變是腦室周圍深靜脈的淤血、擴(kuò)張或梗死的看法不同[4-7]。本研究小組發(fā)現(xiàn)該損傷的分布區(qū)域與神經(jīng)膠質(zhì)細(xì)胞遷移路徑極為相似，在這一路徑上主要分布著向皮層遷移的尚未發(fā)育成熟的少突膠質(zhì)前體細(xì)胞[4]，此細(xì)胞對(duì)興奮性氨基酸的毒作用及缺氧等危險(xiǎn)因素更為敏感而易受損傷[8-9]。這可能是造成該損傷特征性分布的原因，但這一設(shè)想仍需進(jìn)一步探索驗(yàn)證。

目前，國內(nèi)外研究多集中在探測(cè)PWML顯性病灶方面，但是大腦微結(jié)構(gòu)屬性的變化遠(yuǎn)遠(yuǎn)超出顯性病灶的范圍[10-11]。本課題組前期研究結(jié)果表明，PWML伴隨著投射纖維、連合纖維以及聯(lián)絡(luò)纖維等腦白質(zhì)微結(jié)構(gòu)屬性的改變，并且病變類型與其所在位置距離顯性病灶的遠(yuǎn)近程度有關(guān)，反映了發(fā)育進(jìn)程中軸突與少突膠質(zhì)細(xì)胞的營養(yǎng)互助關(guān)系遭到了破壞，從而導(dǎo)致擴(kuò)展性的腦白質(zhì)變性(圖2)，具體體現(xiàn)為少突膠質(zhì)細(xì)胞受損、髓鞘化受阻甚至神經(jīng)元細(xì)胞體受損[8-9,12-13]。雖然PWML被認(rèn)為是一種腦白質(zhì)病，但是本小組研究發(fā)現(xiàn)，PWML伴隨的隱性病變還分布于腦灰質(zhì)區(qū)域，具體表現(xiàn)為腦灰質(zhì)出現(xiàn)不同程度的萎縮，磁共振波譜(MRS) 出現(xiàn)代謝異常，表現(xiàn)為豆?fàn)詈?、丘腦NAA/Cho及丘腦NAA/Cr降低[14]。根據(jù)該變化可以推測(cè)，腦灰質(zhì)區(qū)域微結(jié)構(gòu)屬性也發(fā)生了改變，而微結(jié)構(gòu)屬性的表征將有助于刻畫PWML隱形病變的空間定位，是準(zhǔn)確判斷預(yù)后的基礎(chǔ)。但是，目前尚缺乏對(duì)PWML微結(jié)構(gòu)隱性病變的變化類型及其空間分布的全面深入的研究。

1.2 PWML患兒的個(gè)體化評(píng)估 雖然PWML病變位置具有一定的空間分布規(guī)律，但是從個(gè)體水平的角度觀察，個(gè)體差異十分顯著。個(gè)體差異首先體現(xiàn)在顯性病灶的大小、位置以及分布范圍的廣度等方面；顯性病灶對(duì)功能腦區(qū)結(jié)構(gòu)的影響程度取決于上述因素，從而導(dǎo)致擴(kuò)展性隱性病變的類型與空間分布的差異；針對(duì)發(fā)育中的大腦，個(gè)體差異主要體現(xiàn)在兩個(gè)方面：相同發(fā)育階段不同個(gè)體的差異、不同發(fā)育階段間的差異[15]。PWML患兒處于快速發(fā)育階段，個(gè)體受遺傳與環(huán)境因素的影響又不同，即使相同腦結(jié)構(gòu)的病變也會(huì)導(dǎo)致不同的發(fā)育結(jié)局。個(gè)體化是臨床實(shí)踐工作中的迫切需求，也是直接關(guān)系到每個(gè)患兒疾病診斷與治療的關(guān)鍵環(huán)節(jié)。國內(nèi)外對(duì)個(gè)體化診療問題十分關(guān)注[16-18]，但是，針對(duì)PWML的個(gè)體化診斷鮮見報(bào)道。

圖1 局灶性白質(zhì)損傷患者腦部MRI圖像表現(xiàn)的特點(diǎn)

Fig.1 Characterization of neonatal brain with punctate white matter lesions on MRI images

A：患者女，胎齡34周+5 d，生后9 d。圖示多發(fā)PWML病灶，T1WI呈高信號(hào)，T2WI呈低信號(hào)，ADC圖呈低信號(hào)，SWI幅度圖呈高信號(hào)。 B：患者男，胎齡38周+6 d，生后11 d。箭頭所示PWML病灶，T1WI呈高信號(hào)，T2WI呈低信號(hào)，ADC圖呈低信號(hào)，SWI幅度圖呈低信號(hào)。

圖2 PWML隱性病變的類型

Fig.2 Types of the occult lesions of PWML

1.3 新生兒PWML的預(yù)后評(píng)估 多項(xiàng)研究指出PWML的發(fā)育結(jié)局與其損傷程度密切相關(guān)[19-20]，輕者后期隨訪可無明確的神經(jīng)功能障礙，嚴(yán)重的損傷往往引發(fā)運(yùn)動(dòng)、認(rèn)知障礙等神經(jīng)后遺癥[21]。一旦新生兒PWML發(fā)展形成腦癱，將嚴(yán)重影響整個(gè)家庭的生活質(zhì)量。實(shí)現(xiàn)該類疾病的早期診斷、精準(zhǔn)診斷、早期干預(yù)是減少其危害的主要途徑[22]。

早期對(duì)PWML患兒進(jìn)行預(yù)后判斷是治療措施制定的基礎(chǔ)，但是目前尚無針對(duì)PWML的預(yù)后評(píng)估體系。與國外研究同步，國內(nèi)在新生兒PWML的MRI早期診斷方面積累了大量臨床資料。臨床影像回顧性研究發(fā)現(xiàn)，線狀及混合型PWML病灶較易演變成腦室周圍白質(zhì)軟化癥，臨床出現(xiàn)認(rèn)知障礙、運(yùn)動(dòng)發(fā)育遲緩等神經(jīng)后遺癥[23]。然而，重度PWML常累及皮質(zhì)脊髓束與視輻射等廣泛區(qū)域，與腦癱患兒的微結(jié)構(gòu)改變分布相接近。痙攣性腦癱患兒往往存在感覺、運(yùn)動(dòng)區(qū)白質(zhì)受損，視輻射區(qū)的白質(zhì)損傷亦與早產(chǎn)兒視力相關(guān)[24]。作為運(yùn)動(dòng)與視覺發(fā)育的重要白質(zhì)纖維通路，皮質(zhì)脊髓束與視輻射在發(fā)育早期的損傷中極可能是后續(xù)運(yùn)動(dòng)與視覺功能障礙(腦癱、弱視)的責(zé)任病灶?；谏鲜鰧?duì)PWML患兒的發(fā)育結(jié)局的初步探索可見，建立量化的預(yù)后評(píng)估已具備一定的基礎(chǔ)。但是，需要在全面刻畫PWML顯性與隱性病變的基礎(chǔ)上，建立個(gè)體水平的預(yù)后評(píng)估方法。

總之，目前國內(nèi)外研究現(xiàn)狀表明，臨床對(duì)PWML的認(rèn)知不足，難以滿足臨床實(shí)踐的迫切需求。

2 核心問題的思考

2.1 新生兒腦MRI模板、精細(xì)結(jié)構(gòu)MRI及其腦網(wǎng)絡(luò)分析技術(shù) 新生兒腦MRI模板是圖像分析的基礎(chǔ)。針對(duì)腦發(fā)育研究，國際上已經(jīng)形成了不同年齡段的離散時(shí)間斷面數(shù)據(jù)的腦模板[25-26]，連續(xù)時(shí)間動(dòng)態(tài)腦模板是新生兒腦模板研究的重要發(fā)展方向[27]。新生兒腦結(jié)構(gòu)特殊，發(fā)育進(jìn)程中變化交錯(cuò)、復(fù)雜，觀察PWML相關(guān)灰、白質(zhì)精細(xì)結(jié)構(gòu)的動(dòng)態(tài)變化是預(yù)后評(píng)估的重要環(huán)節(jié)。

新生兒PWML主要體現(xiàn)為腦微結(jié)構(gòu)屬性的變化，擴(kuò)散MRI是刻畫腦組織的微結(jié)構(gòu)特性的重要方法[28-29]。2013年《PNAS》刊文利用擴(kuò)散張量成像(diffusion tensor imaging, DTI)技術(shù)刻畫了早產(chǎn)兒皮層微結(jié)構(gòu)早期發(fā)育的時(shí)空變化規(guī)律[29]。目前，基于腦組織的復(fù)雜特性，描述擴(kuò)散MRI的模型不斷更新，概括來講，擴(kuò)散MRI新技術(shù)主要發(fā)展方向包含以下兩個(gè)方面：①基于不斷完善的水分子擴(kuò)散隨機(jī)過程理論[30-33]；②基于更加精細(xì)的組織結(jié)構(gòu)特性模型(圖3)[34-36]。腦白質(zhì)組織模型和神經(jīng)導(dǎo)向的分散性和密度成像(neurite orientation dispersion and density imaging, NODDI)綜合考慮神經(jīng)元內(nèi)外空間結(jié)構(gòu)特性以及神經(jīng)的方向性，為腦精細(xì)結(jié)構(gòu)的量化提供了更為豐富的指標(biāo)[35-36]。通過對(duì)多變量協(xié)同變化進(jìn)行編碼組合形成不同模式[8]，進(jìn)而反映腦組織結(jié)構(gòu)復(fù)雜的生理或病理變化[37]。結(jié)合腦組織結(jié)構(gòu)屬性模型與多變量協(xié)同變化分析有望揭示PWML顯性與隱性病變微結(jié)構(gòu)空間分布特點(diǎn)。

圖3 基于擴(kuò)散MRI的腦組織結(jié)構(gòu)模型示意圖 Fig.3 Diagram of brain tissue models based on diffusion MRI

新生兒腦網(wǎng)絡(luò)分析是探索早期腦發(fā)育的重要技術(shù)手段?！禤NAS》[38-42]、《Cerebral Cortex》等[43-48]刊出大量應(yīng)用該技術(shù)揭示了腦網(wǎng)絡(luò)屬性的發(fā)育動(dòng)態(tài)變化規(guī)律的報(bào)道。研究新生兒腦網(wǎng)絡(luò)將為認(rèn)識(shí)雙側(cè)半球的對(duì)稱性[49]、模塊化[43]、小世界屬性[46]、RICH-CLUB[39]、默認(rèn)網(wǎng)絡(luò)[38]等腦結(jié)構(gòu)和功能連接出現(xiàn)的時(shí)段及動(dòng)態(tài)變化規(guī)律提供直觀的量化證據(jù)。2010年《Science》發(fā)表的論文利用網(wǎng)絡(luò)連接分析方法用于個(gè)體腦部成熟度的度量[50]。概括來講，應(yīng)用MRI進(jìn)行新生兒腦網(wǎng)絡(luò)分析的方法主要有：DTI腦網(wǎng)絡(luò)[39,45,49,51]、血氧水平依賴(blood oxygen level dependent, BOLD)靜息態(tài)腦網(wǎng)絡(luò)[38,41-42,44,48]、形態(tài)學(xué)指標(biāo)協(xié)變腦網(wǎng)絡(luò)[43,46-47,52]。3類方法得到的腦網(wǎng)絡(luò)側(cè)重點(diǎn)有所不同，DTI腦網(wǎng)絡(luò)與形態(tài)學(xué)指標(biāo)協(xié)變腦網(wǎng)絡(luò)將有助于建立腦精細(xì)結(jié)構(gòu)與腦功能網(wǎng)絡(luò)的關(guān)系，克服“靜息態(tài)”概念中存在的問題，有望在特殊人群鎮(zhèn)靜狀態(tài)下實(shí)現(xiàn)網(wǎng)絡(luò)分析，彌補(bǔ)BOLD在這些方面的不足。

2.2 個(gè)體化分析方法 個(gè)體差異廣泛存在于腦發(fā)育、疾病、認(rèn)知等多個(gè)方面[53-54]。發(fā)育中腦疾病個(gè)體化診斷的關(guān)鍵是建立群體在動(dòng)態(tài)發(fā)育過程中各項(xiàng)量化指標(biāo)的參考模板與置信區(qū)間[55]，該方法也是臨床實(shí)踐中被廣泛應(yīng)用的方法。目前，基于模板向個(gè)體的反向配準(zhǔn)可實(shí)現(xiàn)個(gè)體化的腦區(qū)定位[56]，動(dòng)態(tài)發(fā)育腦模板將在PWML個(gè)體化診斷過程中起重要作用[27,29]。機(jī)器學(xué)習(xí)在該方面具有顯著優(yōu)勢(shì)，通過大樣本的訓(xùn)練，人工智能可以在一定程度上實(shí)現(xiàn)個(gè)體化分析，并被廣泛應(yīng)用于腦齡與行為能力等方面的評(píng)估[18,57]。相關(guān)領(lǐng)域一系列研究的經(jīng)驗(yàn)為PWML個(gè)體化分析提供了基礎(chǔ)。

2.3 早期預(yù)測(cè)方法 基于腦MRI多指標(biāo)項(xiàng)可有效預(yù)測(cè)疾病風(fēng)險(xiǎn)與長(zhǎng)期發(fā)展結(jié)局(包括運(yùn)動(dòng)、感覺甚至認(rèn)知功能)[58-62]。早期預(yù)后評(píng)估涉及到特征提取、篩選、預(yù)測(cè)模型建立與驗(yàn)證等多個(gè)處理環(huán)節(jié)[57]。在常規(guī)MRI提供的灰度、紋理以及各種形態(tài)學(xué)信息的基礎(chǔ)上，擴(kuò)散MRI提供了更加靈敏的反映微結(jié)構(gòu)屬性的參量[63]；影像組學(xué)的研究發(fā)現(xiàn)，納入臨床指標(biāo)項(xiàng)將提高預(yù)測(cè)的準(zhǔn)確性[64-66]。對(duì)于新生兒與嬰幼兒，大腦處于快速發(fā)育階段，發(fā)育因素將使預(yù)測(cè)模型更加復(fù)雜[67-69]。深度學(xué)習(xí)算法使預(yù)測(cè)模型建立的受限條件減少[58]，有助于建立與完善預(yù)測(cè)模型。有學(xué)者基于新生兒期MRI數(shù)據(jù)成功預(yù)測(cè)了腦皮層與白質(zhì)纖維的縱向發(fā)育[70-71]，為下一步的研究提供了技術(shù)保障。

綜上所述，目前仍缺乏新生兒PWML的早期個(gè)體化預(yù)后評(píng)估方案，顱腦MRI與數(shù)據(jù)分析新方法有待轉(zhuǎn)化，融合不同層次的數(shù)據(jù)分析新方法將有助于實(shí)現(xiàn)PWML個(gè)體化預(yù)后評(píng)估。

3 總 結(jié)

《中國兒童發(fā)展綱要(2011-2020年)》將“減少出生缺陷所致殘疾”列為兒童與健康發(fā)展的首要目標(biāo)，并明確指出“促進(jìn)兒童發(fā)展，對(duì)于全面提高中華民族素質(zhì)，建設(shè)人力資源強(qiáng)國具有重要戰(zhàn)略意義”。

綜上所述，PWML病因復(fù)雜，多數(shù)患兒伴有缺血缺氧腦病[8]，同時(shí)伴隨多種并發(fā)癥，其發(fā)病機(jī)制至今尚不明確。值得關(guān)注的是，該類損傷在臨床影像診斷實(shí)踐中發(fā)生率最高，重度可轉(zhuǎn)為腦室周圍白質(zhì)軟化癥(PVL)，導(dǎo)致腦癱；即使是輕度也可見微結(jié)構(gòu)屬性的變化，而且與隱性腦癱患兒微結(jié)構(gòu)病變分布極為相似[72]，二者可能存在潛在的聯(lián)系。本研究組前期研究結(jié)果表明，PWML所致的白質(zhì)結(jié)構(gòu)變化常累及皮質(zhì)脊髓束與視輻射的走行區(qū)域，同時(shí)發(fā)現(xiàn)弱視患兒視輻射DTI參量與皮層厚度顯著相關(guān)[73]。進(jìn)一步提示，作為重要的運(yùn)動(dòng)、視覺通路，皮質(zhì)脊髓束與視輻射區(qū)域可能是PWML相關(guān)腦病的責(zé)任病灶?？梢姡敖沂綪WML患兒腦顯性與隱性病變中的責(zé)任病灶的發(fā)展演變規(guī)律”不僅是預(yù)后判斷的關(guān)鍵，同時(shí)也是認(rèn)識(shí)PWML的發(fā)展機(jī)制的重要組成部分?；陉P(guān)鍵參量與責(zé)任病灶，實(shí)現(xiàn)適用于PWML患兒的個(gè)體化分析，以及綜合考慮個(gè)體差異、發(fā)育因素以及干預(yù)因素，實(shí)現(xiàn)PWML患兒發(fā)育結(jié)局的早期預(yù)測(cè)是該方面臨床研究的核心任務(wù)。

[1] DE BRU?NE FT, VAN DEN BERG-HUYSMANS AA, LEIJSER LM, et al. Clinical implications of MR imaging findings in the white matter in very preterm infants: A 2-year follow-up study[J]. Radiology, 2011, 261(3):899-906.

[2] TORTORA D, PANARA V, MATTEI PA, et al. Comparing 3T T1-weighted sequences in identifying hyperintense punctate lesions in preterm neonates[J]. AJNR Am J Neuroradiol, 2015, 36(3):581-586.

[3] DYET LE, KENNEA N, COUNSELL SJ, et al. Natural history of brain lesions in extremely preterm infants studied with serial magnetic resonance imaging from birth and neurodevelopmental assessment[J]. Pediatrics, 2006, 118(2):536-548.

[4] RAYBAUD C, AHMAD T, RASTEGAR N, et al. The premature brain: Developmental and lesional anatomy[J]. Neuroradiology, 2013, 55(2):23-40.

[5] RUTHERFORD MA, SUPRAMANIAM V, EDERIES A, et al. Magnetic resonance imaging of white matter diseases of prematurity[J]. Neuroradiology, 2010, 52(6):505-521.

[6] PAVLOVA MA, KRGELOH-MANN I. Limitations on the developing preterm brain: Impact of periventricular white matter lesions on brain connectivity and cognition[J]. Brain, 2013, 136(4): 998-1011.

[7] RAETS M, DUDINK J, RAYBAUD C, et al. Brain vein disorders in newborn infants[J]. Dev Med Child Neurol, 2015, 57(3):229-240.

[8] LI X, GAO J, WANG M, et al. Characterization of extensive microstructural variations associated with punctate white matter lesions in preterm neonates[J]. AJNR Am J Neuroradiol, 2017, 38(6):1228-1234.

[9] LI X, GAO J, WANG M, et al. Rapid and reliable tract-based spatial statistics pipeline for diffusion tensor imaging in the neonatal brain: Applications to the white matter development and lesions[J]. Magn Reson Imaging, 2016, 34(9):1314-1321.

[10] BASSI L, CHEW A, MERCHANT N, et al. Diffusion tensor imaging in preterm infants with punctate white matter lesions[J]. Pediatr Res, 2011, 69(6):561-566.

[11] MILLER SP, VIGNERON DB, HENRY RG, et al. Serial quantitative diffusion tensor MRI of the premature brain: Development in newborns with and without injury[J]. J Magn Reson Imaging, 2002, 16(6):621-632.

[12] 高潔，張育苗，孫親利，等. 基于纖維束空間統(tǒng)計(jì)分析法研究早產(chǎn)及足月新生兒局灶性腦白質(zhì)損傷的擴(kuò)散張量成像特點(diǎn)[J]. 磁共振成像, 2014, 5(1):24-29.

[13] ALIZADEH A, DYCK SM, KARIMI-ABDOLREZAEE S. Myelin damage and repair in pathologic CNS: Challenges and prospects[J]. Front Mol Neurosci, 2015, 8(1):35.

[14] SUN Q, YANG J, CHAO J, et al. Detection of ‘silent’ abnormalities in the deep gray matter nuclei of neonates with punctate white matter lesions by magnetic resonance spectroscopy[C]. Proc of RSNA, Chicago, USA, 2017, 17013217.

[15] BROWN TT, KUPERMAN JM, CHUNG Y, et al. Neuroanatomical assessment of biological maturity[J]. Curr Biol, 2012, 22(18):1693-1698.

[16] IYENGAR R, ALTMAN RB, TROYANSKYA O, et al. Personalization in practice[J]. Science, 2015, 350(6258):282-283.

[17] TAVOR I, PARKER JO, MARS RB, et al. Task-free MRI predicts individual differences in brain activity during task performance[J]. Science, 2016, 352(6282):216-220.

[18] DOSENBACH NU, NARDOS B, COHEN AL, et al. Prediction of individual brain maturity using fMRI[J]. Science, 2011, 329(5997):1358-1361.

[19] JEON TY, KIM JH, YOO SY, et al. Neurodevelopmental outcomes in preterm infants: Comparison of infants with and without diffuse excessive high signal intensity on MR images at near-term-equivalent age[J]. Radiology, 2012, 263(2):518-526.

[20] LAURA B, ANDREW C, NAZAKAT M, et al. Diffusion tensor imaging in preterm infants with punctate white matter lesions[J]. Pediatr Res, 2011, 69(6):561-566.

[21] SKI?LD B, VOLLMER B, B?HM B, et al. Neonatal magnetic resonance imaging and outcome at age 30 months in extremely preterm infants[J]. J Pediatr, 2012, 160(4):559-566.

[22] COLVER A, RAPP M, EISEMANN N, et al. Self-reported quality of life of adolescents with cerebral palsy: A cross-sectional and longitudinal analysis[J]. Lancet, 2015, 385(9969):705-716.

[23] 祁英，王曉明. 擴(kuò)散加權(quán)成像在早期診斷早產(chǎn)兒局灶性腦白質(zhì)損傷及其預(yù)后的價(jià)值[J]. 臨床放射學(xué)雜志, 2010, 29(4):515-518.

[24] BASSI L, RICCI D, VOLZONE A, et al. Probabilistic diffusion tractography of the optic radiations and visual function in preterm infants at term equivalent age[J]. Brain, 2008, 131(1):573-582.

[25] OISHI K, MORI S, DONOHUE PK, et al. Multi-contrast human neonatal brain atlas: Application to normal neonate development analysis[J]. Neuroimage, 2011, 56(1):8-20.

[26] SHI F, YAP P-T, WU G, et al. Infant brain atlases from neonates to 1-and 2-year-olds[J]. Plos One, 2011, 6(4):e18746.

[27] SERAG A, ALJABAR P, BALL G, et al. Construction of a consistent high-definition spatio-temporal atlas of the developing brain using adaptive kernel regression[J]. Neuroimage, 2012, 59(3):2255-2265.

[28] WEDEEN VJ, ROSENE DL, WANG R, et al. The geometric structure of the brain fiber pathways[J]. Science, 2012, 335(6076):1628-1634.

[29] GARETH B, LATHA S, PAUL A, et al. Development of cortical microstructure in the preterm human brain[J]. Proc Natl Acad Sci, 2013, 110(23):9541-9546.

[30] JENSEN JH, HELPERN JA, RAMANI A, et al. Diffusional kurtosis imaging: the quantification of non-gaussian water diffusion by means of magnetic resonance imaging [J]. Magn Reson Med, 2005, 53(6): 1432-1440.

[31] BENNETT KM, SCHMAINDA KM, RAOQIONG TONG B, et al. Characterization of continuously distributed cortical water diffusion rates with a stretched-exponential model[J]. Magn Reson Med, 2003, 50(4):727-734.

[32] ZHOU XJ, GAO Q, ABDULLAH O, et al. Studies of anomalous diffusion in the human brain using fractional order calculus[J]. Magn Reson Med, 2010, 63(3):562-569.

[33] FAN Y. Fractional motion model for characterization of anomalous diffusion from NMR signals[J]. Physiol Rev E, 2015, 92(1):012707.

[34] TUCH DS. Q-ball imaging [J]. Magn Reson Med, 2004, 52(6):1358-1372.

[35] FIEREMANS E, JENSEN JH, HELPERN JA. White matter characterization with diffusional kurtosis imaging [J]. Neuroimage, 2011, 58(1):177-188.

[36] HUI Z, SCHNEIDER T, WHEELER-KINGSHOTT CA, et al. NODDI: Practicalinvivoneurite orientation dispersion and density imaging of the human brain[J]. Neuroimage, 2012, 4(4):1000-1016.

[37] DUBOIS J, DEHAENE-LAMBERTZ G, PERRIN M, et al. Asynchrony of the early maturation of white matter bundles in healthy infants: Quantitative landmarks revealed noninvasively by diffusion tensor imaging[J]. Hum Brain Mapp, 2008, 29(1):14-27.

[38] GAO W, ZHU H, GIOVANELLO KS, et al. Evidence on the emergence of the brain’s default network from 2-week-old to 2-year-old healthy pediatric subjects[J]. Proc Natl Acad Sci, 2009, 106(16):6790-6795.

[39] BALL G, ALJABAR P, ZEBARI S, et al. Rich-club organization of the newborn human brain[J]. Proc Natl Acad Sci, 2014, 111(20):7456-7461.

[40] DORIA V, BECKMANN CF, ARICHI T, et al. Emergence of resting state networks in the preterm human brain[J]. Proc Natl Acad Sci, 2010, 107(46):20015-20020.

[41] FRANSSON P, SKIOLD B, HORSCH S, et al. Resting-state networks in the infant brain[J]. Proc Natl Acad Sci, 2007, 104(39):15531-15536.

[42] ZIELINSKI BA, GENNATAS ED, ZHOU J, et al. Network-level structural covariance in the developing brain[J]. Proc Natl Acad Sci, 2010, 107(42):18191-18196.

[43] CHEN ZJ, HE Y, ROSANETO P, et al. Revealing modular architecture of human brain structural networks by using cortical thickness from MRI[J]. Cereb Cortex, 2008, 18(10):2374-2381.

[44] FRANSSON P, ADEN U, BLENNOW M, et al. The functional architecture of the infant brain as revealed by resting-state fMRI[J]. Cereb Cortex, 2011, 21(1):145-154.

[45] HAO H, NI S, MISHRA V, et al. Development of human brain structural networks through infancy and childhood[J]. Cereb Cortex, 2013, 25(5):1389-1404.

[46] HE Y, CHEN ZJ, EVANS AC. Small-world anatomical networks in the human brain revealed by cortical thickness from MRI[J]. Cereb Cortex, 2007, 17(10):2407-2419.

[47] SCHMITT JE, LENROOT RK, WALLACE GL, et al. Identification of genetically mediated cortical networks: A multivariate study of pediatric twins and siblings[J]. Cereb Cortex, 2008, 18(8):1737-1747.

[48] SMYSER CD, INDER TE, SHIMONY JS, et al. Longitudinal analysis of neural network development in preterm infants[J]. Cereb Cortex, 2010, 20(12):2852-2862.

[49] RATNARAJAH N, RIFKIN-GRABOI A, FORTIER MV, et al. Structural connectivity asymmetry in the neonatal brain[J]. Neuroimage, 2013, 75(1):187-194.

[50] DOSENBACH NUF, SCHLAGGAR BL. Prediction of individual brain maturity using fMRI[J]. Science, 2010, 329(5997):1358-1361.

[51] TYMOFIYEVA O, HESS CP, ZIV E, et al. Towards the “baby connectome”: Mapping the structural connectivity of the newborn brain[J]. PLoS One, 2012, 7(2):e31029.

[52] FAN Y, SHI F, SMITH JK, et al. Brain anatomical networks in early human brain development[J]. Neuroimage, 2011, 54(3):1862-1871.

[53] BROWN TT. Individual differences in human brain development[J]. WIREs Cogn Sci, 2017, 8(1):e1389.

[54] HOLMES AJ, HOLLINSHEAD MO, ROFFMAN JL, et al. Individual differences in cognitive control circuit anatomy link sensation seeking, impulsivity, and substance use[J]. J Neurosci, 2016, 36(14):4038-4049.

[55] YEATMAN JD, DOUGHERTY RF, MYALL NJ, et al. Tract profiles of white matter properties: Automating fiber-tract quantification[J]. PLoS One, 2012, 7(11): e49790.

[56] WANG D, BUCKNER RL, FOX MD, et al. Parcellating cortical functional networks in individuals[J]. Nat Neurosci, 2015, 18(12):1853-1860.

[57] SHEN X, FINN ES, SCHEINOST D, et al. Using connectome-based predictive modeling to predict individual behavior from brain connectivity[J]. Nat Protoc, 2017, 12(3):506-518.

[58] HAZLETT HC, GU H, MUNSELL BC, et al. Early brain development in infants at high risk for autism spectrum disorder[J]. Nature, 2017, 542(7641):348-351.

[59] KELLY PJ, ALBERS GW, CHATZIKONSTANTINOU A, et al. Validation and comparison of imaging-based scores for prediction of early stroke risk after transient ischaemic attack: A pooled analysis of individual-patient data from cohort studies[J]. Lancet Neurol, 2016, 15(12):1238-1247.

[60] MILES DK, SMITHERMAN E, HERNANDEZ A, et al. Predicting outcome after pediatric head injury by early magnetic resonance imaging lesion location and volume[J]. J Neurotra, 2016, 33(1):35-48.

[61] DE WIT S, ZIERMANS TB, NIEUWENHUIS M, et al. Individual prediction of long-term outcome in adolescents at ultra-high risk for psychosis: Applying machine learning techniques to brain imaging data[J]. Hum Brain Mapp, 2016, 38(2):704-714.

[62] ROSE J, CAHILLROWLEY K, VASSAR R, et al. Neonatal brain microstructure correlates of neurodevelopment and gait in preterm children 18-22 months of age: An MRI and DTI study[J]. Pediatr Res, 2015, 78(6):700-708.

[63] CHAPMAN CH, TONG Z, NAZEM-ZADEH M, et al. Diffusion tensor imaging predicts cognitive function change following partial brain radiotherapy for low-grade and benign tumors[J]. Radiother Oncol, 2016, 120(2):234-240.

[64] MCGARRY SD, HURRELL SL, KACZMAROWSKI AL, et al. Magnetic resonance imaging-based radiomic profiles predict patient prognosis in newly diagnosed glioblastoma before therapy[J]. Tomography, 2016, 2(3):223-228.

[65] HUANG YQ, LIANG CH, HE L, et al. Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer[J]. J Clin Oncol, 2016, 34(18):2157-2164.

[66] LAMBIN P, RIOS-VELAZQUEZ E, LEIJENAAR R, et al. Radiomics: Extracting more information from medical images using advanced feature analysis[J]. Eur J Cancer, 2012, 48(4):441-446.

[67] CHERUBINI A, CALIGIURI ME, PERAN P, et al. Importance of multimodal MRI in characterizing brain tissue and its potential application for individual age prediction[J]. IEEE J Biomed Health Inform, 2016, 20(5): 1232-1239.

[68] LIEM F, VAROQUAUX G, KYNAST J, et al. Predicting brain-age from multimodal imaging data captures cognitive impairment[J]. Neuroimage, 2017, 148(1):179-188.

[69] EWING-COBBS L, JOHNSON CP, JURANEK J, et al. Longitudinal diffusion tensor imaging after pediatric traumatic brain injury: Impact of age at injury and time since injury on pathway integrity[J]. Hum Brain Mapp, 2016, 37(11):3929-3945.

[70] MENG Y, LI G, REKIK I, et al. Can we predict subject-specific dynamic cortical thickness maps during infancy from birth?[J]. Hum Brain Mapp, 2017, 38(6):2865-2874.

[71] REKIK I, LI G, PEW-THIAN Y, et al. Joint prediction of longitudinal development of cortical surfaces and white matter fibers from neonatal MRI[J]. NeuroImage, 2017, 152(1): 411-424.

[72] MU X, NIE B, WANG H, et al. Spatial patterns of whole brain grey and white matter injury in patients with occult spastic diplegic cerebral palsy[J]. PLoS One, 2014, 9(6):e100451.

[73] QI S, MU YF, CUI LB, et al. Association of optic radiation integrity with cortical thickness in children with anisometropic amblyopia[J]. Neurosci Bull, 2016, 32(1):1-10.