-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Many patients with drug-refractory temporal lobe epilepsy (TLE) have hippocampal sclerosis (HS)
.
MRI helps to identify such lesions that may form epileptic foci, thereby simplifying pre-operative evaluation
epilepsy
The main imaging feature of hippocampal sclerosis is the loss of hippocampal volume, which is usually related to low-density T1 and high-density T2 weighted signals
.
The biological validity of these features has been confirmed by the combined analysis of MRI and histopathology, indicating that cell density reduction and glial hyperplasia are positively correlated with atrophy and T2 hyperenhancement, respectively
However, in clinical practice, MRI still cannot show hippocampal lesions in 30-50% of surgical candidates with clear clinical electrical evidence of TLE
.
This wide range may be partly due to suboptimal imaging protocols and limited professional experience
In addition, although quantitative analysis, including hippocampal volume measurement, voxel-based morphometry, T2 relaxometry, and FLAIR signal intensity measurements have proven to be more sensitive than visual assessments, they are still underutilized
.
It is worth noting that the in vivo characteristics of HS are affected by the severity of neuron loss and glial hyperplasia.
Therefore, many patients who are not shown by MRI may have to undergo an intracranial EEG examination.
The risks of this examination are similar to those of resection surgery, and the cost is high
.
Although there is a large amount of neuroimaging literature to assess the structural integrity of the hippocampus of TLE, the vast majority of studies are aimed at group-level changes
.
On the other hand, individual analysis usually uses a single comparison to normalize it to the distribution of healthy subjects, and rarely solves the focus of MRI-negative patients
Since the typical features of HS are T1-weighted low density and T2-weighted high density, they also generate a composite contrast by dividing the FLAIR intensity by the T1-weighted intensity, thereby maximizing their comprehensive contribution to detect the complete HS spectrum
.
We applied this classifier to the MRI features of HS in TLE and evaluated its generality in two independent cohorts
They trained a surface-based linear discriminant classifier, using T1 weighting (morphology) and T2 weighting and FLAIR/T1 (intensity) features
.
.
The classifier was trained on 60 TLE patients (mean age: 35.
6 years; 58% women) who had histologically confirmed hippocampal sclerosis (HS)
The predictive model automatically marks the patient as left or right TLE
.
Lateralization accuracy was compared with EEG-clinical data, including the side of surgery
.
The accuracy of the classifier was further evaluated in two independent TLE cohorts with similar demographic and electroclinical characteristics (n=57; 58% MRI negative)
.
They found that regardless of the visibility of HS on MRI, the overall lateralization accuracy rate was 93% (95%; CI 92%-94%)
.
In MRI-negative TLE, the combination of T2 and FLAIR/T1 intensities was used in training (84%, area under the curve (AUC): 0.
95±0.
02) and validation cohort (cohort 1: 90%, AUC: 0.
99; cohort 2: 76 %, AUC: 0.
94) provides the highest accuracy
.
The model predicted lateral TLE is to ready-made conventional MRI contrast based and provides more accuracy than visual and radiographic evaluation
.
The combined contribution of the decrease of T1 weighted intensity and the increase of T2 weighted intensity makes the synthetic FLAIR/T1 contrast particularly effective for MRI-negative HS, laying the foundation for a wide range of clinical transformations
.
Original source:
Caldairou B, Foit NA, Mutti C, et al.
An MRI-Based Machine Learning Prediction Framework to Lateralize Hippocampal Sclerosis in Patients With Temporal Lobe Epilepsy.
Neurology.
Published online September 2, 2021:10.
1212/WNL.
0000000000012699.
doi :10.
1212/WNL.
0000000000012699
