1naresh
Array
(
    [urn:ac.highwire.org:guest:identity] => Array
        (
            [runtime-id] => urn:ac.highwire.org:guest:identity
            [type] => guest
            [service-id] => ajnr-ac.highwire.org
            [access-type] => Controlled
            [privilege] => Array
                (
                    [urn:ac.highwire.org:guest:privilege] => Array
                        (
                            [runtime-id] => urn:ac.highwire.org:guest:privilege
                            [type] => privilege-set
                            [privilege-set] => GUEST
                        )

                )

            [credentials] => Array
                (
                    [method] => guest
                )

        )

)
1naresh
Array
(
    [urn:ac.highwire.org:guest:identity] => Array
        (
            [runtime-id] => urn:ac.highwire.org:guest:identity
            [type] => guest
            [service-id] => ajnr-ac.highwire.org
            [access-type] => Controlled
            [privilege] => Array
                (
                    [urn:ac.highwire.org:guest:privilege] => Array
                        (
                            [runtime-id] => urn:ac.highwire.org:guest:privilege
                            [type] => privilege-set
                            [privilege-set] => GUEST
                        )

                )

            [credentials] => Array
                (
                    [method] => guest
                )

        )

)


Characteristics of the multiple linear regression model for predicting limb strength from clinical and MR imaging data
Ankle Dorsiflexion (lb)Hip Flexion (lb)
Medulla IncludedMedulla ExcludedMedulla IncludedMedulla Excluded
Number of ankles*72886883
Model coefficients†
    Midbrain, median MTR−276 ± 86 (P = .002)−224 ± 78 (p = .005)−183 ± 68 (P = .01)−158 ± 71 (P = .03)
    Pons, median MTR300 ± 99 (P = .003)254 ± 83 (P = .003)175 ± 88 (P = .03)176 ± 75 (P = .004)
    Medulla, median MTR153 ± 57 (P = .009)134 ± 48 (P = .006)
    MS clinical subtype‡−21.7 ± 3.9 (P < .001)−19.5 ± 3.7 (P < .001)−22.9 ± 3.2 (P < .001)−20.4 ± 3.3 (P < .001)
    Constant−42 ± 59 (P = .5)−26 ± 48 (P = .6)−21 ± 47 (P = .7)−27 ± 43 (P = .5)
Partial correlation coefficients§
    Midbrain, median MTR−0.37 (P = .002)−0.30 (P = .005)−0.32 (P = .01)−0.24 (P = .03)
    Pons, median MTR0.35 (P = .003)0.32 (P = .003)0.27 (P = .03)0.26 (P = .02)
    Medulla, median MTR0.31 (P = .009)0.34 (P = .006)
    MS clinical subtype−0.56 (P < .001)−0.50 (P < .001)−0.67 (P < .001)−0.57 (P < .001)
Model performance
    Adjusted r20.36 (P < .0001)0.30 (P < .0001)0.45 (P < .0001)0.33 (P < .0001)
    Predictions within 10% of actual strength28%21%5%30%
    Median difference of prediction from actual strength13%21%50%17%
Model performance vs prior imaging and strength testing
    r20.34 (P < .0001)0.17 (P = .0004)0.38 (P < .0001)0.28 (P < .0001)
    Predictions within 10% of actual strength24%15%25%22%
    Median difference of prediction from actual strength21%23%32%29%
  • * Each subject contributes up to 2 ankles (right and left). The number of hips and ankles listed here is lower than the overall total because MTR was not obtained for all subjects. The model does not account for correlations between ankles in the same subject. Pearson correlation coefficients are used.
  • † Coefficient errors are ± 1 standard deviation.
  • ‡ MS clinical subtype is modeled as 1 for secondary progressive MS, 0 otherwise.
  • § Pearson correlation coefficients for model parameter, holding the others constant.