Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade II glioma

Abstract

Objective

The status of isocitrate dehydrogenase 1 (IDH1) is highly correlated with the development, treatment and prognosis of glioma. We explored a noninvasive method to reveal IDH1 status by using a quantitative radiomics approach for grade II glioma.

Methods

A primary cohort consisting of 110 patients pathologically diagnosed with grade II glioma was retrospectively studied. The radiomics method developed in this paper includes image segmentation, high-throughput feature extraction, radiomics sequencing, feature selection and classification. Using the leave-one-out cross-validation (LOOCV) method, the classification result was compared with the real IDH1 situation from Sanger sequencing. Another independent validation cohort containing 30 patients was utilised to further test the method.

Results

A total of 671 high-throughput features were extracted and quantized. 110 features were selected by improved genetic algorithm. In LOOCV, the noninvasive IDH1 status estimation based on the proposed approach presented an estimation accuracy of 0.80, sensitivity of 0.83 and specificity of 0.74. Area under the receiver operating characteristic curve reached 0.86. Further validation on the independent cohort of 30 patients produced similar results.

Conclusions

Radiomics is a potentially useful approach for estimating IDH1 mutation status noninvasively using conventional T2-FLAIR MRI images. The estimation accuracy could potentially be improved by using multiple imaging modalities.

Key Points

• Noninvasive IDH1 status estimation can be obtained with a radiomics approach.

• Automatic and quantitative processes were established for noninvasive biomarker estimation.

• High-throughput MRI features are highly correlated to IDH1 states.

• Area under the ROC curve of the proposed estimation method reached 0.86.

Appendices

Appendix 1

A genetic algorithm (GA) usually starts with a population of chromosomes each of which represents a candidate solution to optimising a problem. The fitness of each chromosome is evaluated through an objective function. In following an optimisation procedure, a stronger chromosome has a higher probability of being selected. Selected chromosomes are subjected to crossover and mutation. Crossover and mutation are performed by exchanging portions of chromosomes and changing parts of the chromosome string, respectively. When the selection, crossover and mutation are iterated at certain times, the fittest chromosome is chosen to solve the current optimisation problem.

In the original GA, the fitness of a chromosome is evaluated by the accuracy of the classification, which can be set out as:

$$ Fitness(c)= Accuracy(c) $$

(1)

where c represents a chromosome. The fitness evaluation in (10) does not take the mutual relationship between chromosomes into consideration, which leads to a relatively narrower selection space and higher dimensions of the selected feature space. A new objective function is therefore used to evaluate the fitness of a chromosome:

$$ Fitness\hbox{'}(c)= Accuracy(c)+1\hbox{-} \frac{Rank(c)}{1+ Accuracy(c)} $$

(2)

where Rank(c) represents the sum of mRMR [25] order number.

Appendix 2

If true positive, true negative, false positive and false negative are represented by TP, TN, FP and FN, respectively, accuracy (ACC), sensitivity (SENS), specificity (SPEC), positive predictive value (PPV), negative predictive value (NPV) and Matthew’s correlation coefficient (MCC) can be calculated as follows:

$$ ACC=\frac{TP+TN}{TP+TN+FP+FN} $$

(3)

$$ SENS=\frac{TP}{TP+FN} $$

(4)

$$ SPEC=\frac{TN}{TN+FP} $$

(5)

$$ PPV=\frac{TP}{TP+FP} $$

(6)

$$ \mathrm{N}\mathrm{P}\mathrm{V}=\frac{TN}{TN+FN} $$

(7)

$$ MCC=\frac{TP\times TN-FP\times FN}{\sqrt{\left(TP+FP\right)\left(TP+FN\right)\left(TN+FP\right)\left(TN+FN\right)}} $$

(8)

For an improved GA method, the population size is set as 50, generation number at 30, crossover probability at 0.9 and mutation probability at 0.1.

Appendix 3

Table 6

Table 6 The following table summarises the shape and texture features with statistical significance for IDH1 differentiation

Table 7

Table 7 The following table only presents the values of 38 wavelet features with p < 0.001. In the table, the grey level bar on the left side denotes features from different wavelet decompositions. LLL, HLL, LHL, HHL, LLH, HLH, LHH and HHH are represented by the grey level in shading from brighter to darker. The symbols #” and *represent features of texture and intensity, respectively