Table 7.

Performance of heterogeneous echo target tasks using the proposed method and the baseline. Bold values indicate best performance.

Method	Tuning Strategy	Target Task	Accuracy	F-score	IoU	MCC
	Independent	Envelope Segmentation	0.97	0.95	0.98	–
	Alternating	Envelope Segmentation	0.96	0.94	0.98	–
	Joint	Envelope Segmentation	0.91	0.93	0.94	–
	Independent	Quality Assessment	0.97	0.96	–	0.92
${\text{UMS-Rep}}_{echo}$	Alternating	Quality Assessment	0.95	0.91	–	0.87
	Joint	Quality Assessment	0.93	0.88	–	0.83
	Independent	Flow Classification	0.93	0.91	–	0.85
	Alternating	Flow Classification	0.91	0.90	–	0.82
	Joint	Flow Classification	0.92	0.90	–	0.84
Baseline	Separate Model	Envelope Segmentation	0.93	0.92	0.95	–
	Separate Model	Quality Assessment	0.88	0.85	–	0.78
	Separate Model	Flow Classification	0.91	0.88	–	0.81